Global Factor Data Documentation

Theis Ingerslev Jensen

Bryan Kelly
Lasse Heje Pedersen*

∗
Jensen is at Yale School of Management; https://sites.google.com/view/theis-ingerslev
-jensen/. Kelly is at AQR Capital Management, Yale School of Management, and NBER; www
.bryankellyacademic.org. Pedersen is at AQR Capital Management, Copenhagen Business School, and
CEPR; www.lhpedersen.com. We are grateful to Faheem Almas and Tyler Gwinn for excellent research
assistance. AQR Capital Management is a global investment management firm, which may or may not
apply similar investment techniques or methods of analysis as described herein. The views expressed here
are those of the authors and not necessarily those of AQR.
Table of Contents
1 Overview 2
1.1 How To Run the Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 How To Use the Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Versions, Bug Fixes, and Comments . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2 Factor Portfolio Construction 7

3 Identifier Variables 7

4 Industry Identification 8
4.1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

5 Helper Functions 9

6 Accounting Characteristics 10
6.1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6.2 General Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6.3 Annualized Accounting Variables from Quarterly Data . . . . . . . . . . . . 10
6.4 Accounting Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

7 Market Based Characteristics 25

8 Detailed Characteristic Construction 30

9 FX Conversion Rate Construction 36

10 Factor Details and Citations 37

11 Miscellaneous 48

References 50

1
1 Overview
• This documentation describes the Global Factor Data, and the associated code for
constructing the data, based on Jensen, Kelly, and Pedersen (2022). The citation for
use of this data and code is:
@article{JensenKellyPedersen2022,
author = {Jensen, Theis Ingerslev and Kelly, Bryan T and Pedersen, Lasse Heje},
journal = {Journal of Finance, Forthcoming},
title = {Is There A Replication Crisis In Finance?},
year = {2022}
}
• The Global Factor Data includes 406 characteristics and their associated factor port-
folios. This is a superset of the 153 factors analyzed in Jensen, Kelly, and Pedersen
(2021).
• This documentation is grouped into eight main sections: Identifier Variables, Industry
Identifiers, Helper Functions, Accounting Characteristics, Market Based Character-
istics, Detailed Characteristic Construction, FX Conversion Rate Construction and
Factor Details and Citations.
– Identifier Variables include firm identifying information, date, etc...
– Each of the Characteristics sections includes at least three subsections: Datasets,
Variables, and Characteristics.
– Datasets refers to which datasets the items in variables are drawn from. For
example, ’COMP.FUNDA’ suggests we use variables from the FUNDA dataset
provided by Compustat.
– Variables refers to a table containing information about the variables drawn from
the datasets previously identified. These tables include the name, abbreviation
used throughout the section, and the construction of the variables. These variables
are constructed in a way to maximize coverage and are not directly included in
the final dataset.
– Characteristics refers to a table of constructed characteristics made of the previ-
ously describes variables. These tables include the name, the abbreviation used
in the published dataset, and the construction. These characteristics are in the
final dataset.

1.1 How To Run the Code

• Access the code for this data set at https://github.com/bkelly-lab/ReplicationCrisis.
• This data is produced using the SAS Studio on Wharton Research Data Services
(WRDS) servers. The github README file contains instruction on how to gener-
ate the data.
• Use the ’EOM’ (end of month) variable as the date variable to join/merge datasets.

2
1.2 How To Use the Data
• The id column is the unique security × source1 identifier.

• The eom column shows the end of month, where the data is valid. In other words, it
shows the information available by the end of a given month. As an example, the me
value for a stocks with eom=20191231, will be the last available market equity before
or at December 31st 2019. When creating portfolio, characteristics should therefore be
lagged relative to returns to avoid lookahead bias.

• The excntry column, identifies the country of the exchange where the security is
traded.

• Suggested screens:

– To restrict the sample to one observation per security × month, use obs main=1.
– To restrict the sample to common stocks, use common=1.
– To restrict the sample to prominent exchanges, use exch main=1.
– To restrict the sample to primary listing × month, use primary sec=1.

1.3 Versions, Bug Fixes, and Comments

• We will update the code and data regularly as CRSP and Compustat updates become
available. We will also release periodic updates with bug fixes. Our GitHub repository:
https://github.com/bkelly-lab/ReplicationCrisis/blob/master/GlobalFactors/
CHANGELOG.md tracks the evolution in the code used to generate the data. Furthermore,
Table 1 gives an overview of the most important changes.

• The code and data has been carefully vetted, but may contain bugs and certainly has
room for improvement. We welcome any and all feedback regarding bugs or suggestions
for improvements and extensions.

• Send correspondence to [email protected] with subject “Global Factor Data”

Table 1: Log of Important Code and Methodology

Date Changes
03-03-2023
• Added ’me’ (market equity) and ’ret’ (total return) and removed
’source crsp’ from daily return files.

1
CRSP or Compustat.

3
Date Changes
02-08-2022
• Fixed error in the construction of intrinsic value. Previously, we failed
to scale intrinsic value by market equity as done in Frankel and Lee
(1998). We call the new characteristic ival me and keep intrinsic value
in the data set. The alpha of the new factor based on ival me is signif-
icantly different from zero, while the factor based on intrinsic value is
insignificant.

11-16-2021
• Changed return cutoffs to depend on all stocks, instead of only stocks
from CRSP.
• Changed the ’source’ (character) column to ’source crsp’ (integer),
source crsp is 1 if CRSP is the return data source.
• Changed the ’id’ column from character to integer. For stocks from
CRSP, the id is just their permno. For stocks from Compustat, the
first digit is 1 if the stock is traded on a US exchange, 2 if it’s traded
on a Canadian exchange, and 3 otherwise. The next two digits are the
IID from Compustat, and the remaining six digits are the gvkey.
• Adapted the primary sec column such that all observations from CRSP
have primary sec=1.
• Changed the treatment of zero return. Previously, we treated a zero
return as a missing observation. Now, we have removed this screen,
such that a zero return is treated like any other return.
• Changed the creation of characteristics based on daily stock market
data. Previously, we winsorized daily returns, market equity, and dol-
lar volume, before creating characteristics based on daily stock market
data. Now, we have removed this winsorization, and daily characteris-
tics are based on the raw data.
• Added the option to create daily factor return in the portfolios.R code.
• Added the option to create industry returns in the portfolios.R code.

08-27-2021
• Fixed a bug regarding how daily delisting returns from CRSP are in-
corporated.
• Added indfmt=’FS’ to the international accounting data.

4
Date Changes
06-14-2021
• We changed the winsorization scheme. First, we removed the
0.01%/99.9% winsorization of market equity in all countries. Second,
we removed the winsorization of returns from the CRSP database. For
Compustat returns, we set returns above (below) the 99.9% (0.01%)
of CRSP returns in the same month, to that level. In other words,
we base our winsorization of Compustat data on CRSP data from the
same month.

02-19-2021
• Previously we did not exclude securities that are only traded over the
counter. In the new version of the data set, we include an indicator
column ”exch main” to exclude non-standard exchanges. In the US,
the main exchanges are AMEX, NASDAQ, and NYSE. Outside of the
US, we exclude over the counter exchanges, stock connect exchanges in
China, and cross-country exchanges such as BATS Chi-X Europe. The
documentation includes a full list of the excluded exchanges.
• Included SIC, NAICS, and GICS industry codes.

02-15-2021
• Removed a bug that caused ivol ff3 21d, iskew ff3 21d, ivol hxz4 21d,
and iskew hxz4 21d to require 17 (ff3) and 18 (hxz) observations for a
valid estimate. Consistent with our original intent, we now require at
least 15 observations for a valid estimate.

02-01-2021
• Fixed a small bug in the bidask hl() macro.
• Lowered the requirement for the number of stocks needed when cre-
ating asset pricing factors (FF and HXZ). We previously required at
least 5 stocks in a sub-portfolio (e.g., small stocks with high BM) for
the observation to be valid. This led to missing observations in the
1950s for small stocks with low BM. We lowered this requirement to 3
stocks. Furthermore, when creating asset pricing factors, we changed
the breakpoints to be based on NYSE stocks in the US instead of non-
microcap stocks. Outside of the US, breakpoints are still based on
non-microcap stocks.

5
Date Changes
01-25-2021
• Changed residual momentum characteristics (resff3 12 1 & resff3 6 1)
to be scaled with the standard deviation of residuals consistent with
Blitz, Huij, and Mertens (2011).
• Fixed error in creating qmj prof. The issue was that the oaccruals at
used the value instead of the z-score of ranks. This effectively meant
that accruals didn’t impact the profitability score.
• Fixed error for annual seasonality characteristics (factor names starting
with seas and ending with an). There was a bug in the screening
procedure which meant that the characteristic for one stock could use
information from an unrelated stock.
• Rounding issues when converting a .csv file to an excel file, caused
the zero trades * variables to not have any decimals which made the
turnover tie-breaker ineffective.
• Standardized unexpected earnings (niq su) and sales (saleq su) is com-
puted as the actual value minus the expected value (standardized by
the standard deviation of this change). Before, the expected value was
computed as the mean yearly change over the last 8 quarters added
to the last quarterly value. Now the expected value is the same mean
yearly change, but added to the quarterly value 4 quarters ago consis-
tent with Jegadeesh and Livnat (2006).

1.4 Terminology
• Annual data refers to accounting data from annual reports sourced from COMP.FUNDA
and COMP.G FUNDA.

• Quarterly data refers to accounting data from quarterly reports sourced from COMP.FUNDQ
and COMP.G FUNDQ.

• Final Dataset refers to “world data.sas7bdat”, the output dataset

• Fiscal period refers to the relevant period over which income and expenses have accrued.

• Accounting variables refers to accounting items such as assets, sales and net income.

• Market variables refers to market based items such as market equity and excess return.

• Characteristics refers to columns in the final dataset that reveals a characteristic about
the security, For example asset growth, book to market equity, and net income to book
equity.

6
2 Factor Portfolio Construction
• For each characteristic, we build the 1-month holding period factor return within each
country as follows.

• In each country and month, we sort stocks into characteristic terciles (top/middle/bot-
tom third) with breakpoints based on non-micro stocks in that country. Specifically, we
start with all non-micro stocks in a country (i.e., larger than NYSE 20th percentile) and
sort them into three groups of equal numbers of stocks based on the characteristic, say
book-to-market. Then we distribute the micro-cap stocks into the three groups based
on the same characteristic breakpoints. This process ensures that the non-micro stocks
are distributed equally among across portfolios, creating more tradable portfolios.

• For each tercile, we compute its “capped value weight” return, meaning that we weight
stocks by their market equity, winsorized at the NYSE 80th percentile. This construc-
tion ensures that tiny stocks have tiny weights and any one mega stock does not
dominate a portfolio, seeking to create tradable, yet balanced, portfolios.

• The factor is then defined as the high-tercile return minus the low-tercile return, corre-
sponding to the excess return of a long-short zero-net-investment strategy. The factor
is long (short) the tercile identified by the original paper to have the highest (lowest)
expected return.

• For a factor return to be non-missing, we require that it has at least 5 stocks in each of
the long and short legs. We also require a minimum of 60 valid monthly observations
for each country-specific factor for inclusion in our sample.

• We update characteristics with the most recent accounting data (which could be either
annual or quarterly) starting four months after the end of the fiscal period.

• To compute a cluster (theme) return, we first sign factors according to the original
reference, then we equal-weight the returns of factors within a specific cluster. The
signing convention and cluster allocation follows Jensen et al. (2022) and we show it
in table 9.

3 Identifier Variables
This section covers all of the variables that give firm/date level identifiers and information.
If a variable starts with ’comp’ or ’crsp’, then the following variable name is drawn from the
specified dataset. For example, ’crsp shrcd’ is the ’shrcd’ variable from CRSP.
Table 2: Identifier Variables
Name Description
This groups each firm into one of five categories: Mega, large, small, micro and nano cap. The groups are
non-overlapping and the breakpoints are based on the market equity of NYSE stocks by the end of each
size grp month. In particular, Mega caps are stocks with a market cap above the 80th percentile of NYSE stocks,
large caps are all remaining stocks above the 50th percentile, small caps are above the 20th percentile, micro
caps above the 1st percentile and nano caps are the remaining stocks.

7
 Name Description
We generate a unique number for each security in our data set. For securities from CRSP, the id is just
the corresponding permno. For stocks from Compustat, the first digits is 1 if the stocks is traded on a US
id
exchange, 2 if it’s traded on a Canadian exchange, and 3 otherwise. The next six digits are the gvkey and the
last two are the iid.2
source crsp Identifies the source of the return data. A 1 (0), indicates that the source is CRSP (Compustat).
For US stocks, we often have two observations for each security-month pair. One from Compustat, and one
from CRSP. In cases with duplicates, the observation from CRSP has obs main=1, and the observation from
obs main Compustat has obs main=0. If there are more than one firm observations for one date, this identifies if the
observation is considered as the ’main’ observation. If available, CRSP observations are considered as the
’main’ observation.
Indicator for ordinary exchanges. If CRSP is the source, main exchanges are those with crsp exchcd 1, 2 and
exch main 3. If Compustat is the source, main exchanges are all comp exchg except 0, 1, 2, 3, 4, 13, 15, 16, 17, 18, 19,
20, 21, 127, 150, 157, 229, 263, 269, 281, 283, 290, 320, 326, 341, 342, 347, 348, 349, 352.
gvkey Permanent six-digit unique firm identifier from Compustat
iid Permanent two-digit addition to ’gvkey’ that identifies specific security of a firm from Compustat
Primary security as identified by Compustat. A ’gvkey’ can have up to three different primary securities (’iid)’
primary sec
at a given time (US, CA, and international). All observations from CRSP has primary sec=1.
permco Permanent unique firm identifier from CRSP
permno Permanent security identifier from CRSP
The country of the exchange where the security is traded. Usually expressed as an ISO currency code with
excntry
the exception of mul which indicates a multi country exchange3
curcd Currency of prc local and the currency used to calculated ret local.
fx Ratio of curcd to USD at the date of observation
Indicator for common stocks. If CRSP is the source, common is one if the SHRCD variable is 10, 11 or 12. If
common
Compustat is the source, common is one if TPCI is ’0’
comp tpci Compustat issue type identifier
crsp shrcd CRSP share code
comp exchg Compustat stock exchange code
crsp exchd CRSP stock exchange code
date Date of the last return observation during the month.
eom The last day of the month in which the observation is made
adjfct Share adjustment factor, using ’cfacshr’ if the source is CRSP or ’ajexdi’ if the source is Compustat

4 Industry Identification
This section describes the industry identifiers. First we contruct separate identifiers for
CRSP and Compustat. Based on these datasets, we create one SIC, NAICS and GICS code
for each firm based on Compustat data if available and otherwise CRSP. GVKEY is the
company identifier for COMPUSTAT. PERMNO is the security identifier for CRSP. While
we would prefer to use PERMCO, which is company level, different firms with different in-
dustry identifications can be listed under the same PERMCO. CRSP identifiers are available
on a daily basis. For Compustat, we extract SIC and NAICS codes from annual accounting
reports. Historical GICS codes are only available in Compustat. The Fama-French industry
identifier is mapped from SIC codes using documentation provided by Ken French. We allow
for using either 38 or 49 industry portfolio definitions, as defined here and here, respectively.
By default, we use the 49 portfolio definition, but that can be adjusted in ’main.sas’.
2
In Compustat, a security is identified by gvkey and iid. To map our id to Compustat, add ’C’ or ’W’ to
the iid if the first digit is 2 or 3 respectively.
3
Typically over the counter exchanges.

8
4.1 Datasets
• CRSP.DSENAMES

• COMP.FUNDA

• COMP.G FUNDA

• COMP.CO HGIC

• COMP.G CO HGIC

Table 3: Identifier Variables

Name Description
sic Firm SIC industry. We use Compustat data if available and otherwise use CRSP data.
naics Firm NAICS industry. We use Compustat data if available and otherwise use CRSP data.
gics Firm GICS industry. We use historical data from Compustat.
Classification of stocks into 49 industry groups based SIC codes and the methodology in Fama and French
ff49
(1997) with the addition of a software industry.

5 Helper Functions
This section describes functions that we use to create variables. Many of the functions are
used for variables with quarterly, monthly and daily frequencies, and these are specified
by “ zQ”, “ zM” and “ zD” respectively, where “z” is the number of quarters, months or
days that the function is referencing. For example, COVAR 12M(X, Y) is the covariance of
variables X and Y over the past 12 months.

Table 4: Helper Functions

Function Name Description

1 Pz−1
Mean Xz z n=0 Xt−n
1 Pz−1
Variance VARC z(X) z−1 n=0 (Xt−n − Xtz )2
1 Pz−1
Covariance COVAR z(X, Y) z−1 n=0 (Xt−n − Xtz )(Yt−n − Ytz )
p
Standard Deviation σz (X) V ARC z(X)
1 Pz−1
Skewness SKEW z(X) z×σz (X)3 n=0 (Xt−n − Xtz )3
Xt −(Xt−3 +(Xt−3 −Xt−15 )z /4)
Standardized Unexpected Realization SUR z(X) σz (Xt−3 −Xt−15 )

Xt
Change to Expectations CHG TO EXP(X) (Xt−12 +Xt−24 )/2

Maximum MAXn z(X) The maximum n values of given input.

Quality Minus Junk Helpers

ROEQ BE ST D × 2. If this is unavailable, we use

Earnings Volatility EV OL
ROE BE ST D.

9
 Function Name Description

Rank of Variable rV ar Cross-sectional rank of Var within a country4

rV AR− rV ARz
Z transformation ZV (rV ar) t( rV AR)

6 Accounting Characteristics
6.1 Datasets
• COMP.FUNDA

• COMP.FUNDQ

• COMP.G FUNDA

• COMP.G FUNDQ

6.2 General Information

• We create characteristics for annual and quarterly accounting data separately. We then
take the most recent characteristics value from each dataset to create the final dataset.

• We assume that accounting variables are publically available 4 months after the end
of the accounting period.

• In describing accounting variables, we use the Compustat item names from the annual
dataset. The equivalent item name in the quarterly dataset can be found by adding
a ‘q’ or ‘y’ to the end of the annual item name. Specifically, ‘q’ indicates a value
calculated over one quarter while ‘y’ refers to the cummulative value over the quarters
with data available within a fiscal year.

6.3 Annualized Accounting Variables from Quarterly Data

• The value of a balance sheet item such as asset or book equity has the same meaning
in the annual and the quarterly data. It is the value by the end of a fiscal period.

• The value of an income or cash flow statement item is different. In the annual data, it is
calculated over one year. However, in the quarterly data, it is calculated over one quar-
ter. To make quarterly income and cash flows items comparable to the corresponding
annual item, we take the sum of the item over the last four quarters.
4
OACCRU ALS AT , BET ABAB 1260d, DEBT AT and EV OL are sorted in descending order. All
other variables are sorted in ascending order.

10
6.4 Accounting Variables
The abbreviation is used to refer to the accounting variable. A suffix of ’*’ indicates that
we have altered the original Compustat item to increase the coverage or to create a variable
that is a part of creating a characteristic in the final dataset. The characteristic name will
reflect the accounting name except the ’*’ suffix. As an example, ’gp at’ is gross profit scaled
by assets. In general, we will refer to Compustat variables using capital letters.
Table 5: Accounting Variables
Name Abbreviation Construction
Income Statement
Sales sale* We prefer SALE. If this is unavailable, we use REVT
Cost of Goods Sold cogs Compustat item COGS
Gross Profit gp* We prefer to use GP. If this is unavailable we use sale*-COGS
Selling, General and Administrative
xsga Compustat item XSGA
Expenses
Compustat item XAD. Note that this is not available in Com-
Advertising Expenses xad
pustat Global
Compustat item XRD. Note that this is not available in Com-
Research and Development Expenses xrd
pustat Global
Staff Expenses xlr Compustat item XLR
Special Items spi Compustat item SPI
We prefer to use XOPR. If this is unavailable, we use
Operating Expenses opex*
COGS+XSGA
We prefer to use EBITDA. If this is unavailable, we use
Operating Income Before Depreciation ebitda* OIBDP. If this is unavailable, we use SALE*-OPEX*. If this
is unavailable, we use GP*-XSGA
Depreciation and Amortization dp Compustat Item DP
We prefer to use EBIT. If this is unavailable, we use OIADP.
Operating Income After Depreciation ebit*
If this is unavailable, we use EBITDA*-DP
Interest Expenses int Compustat item XINT
We use EBITDA* + XRD. If XRD is unavailable, we set it to
Operating Profit ala Ball et al (2015) op*
zero
We use EBITDA*-XINT. Note that we target the same vari-
able as the numerator of the profitability characteristic used
Operating Profit to Equity ope*
to create the Robust-minus weak factor in the fama-French 5
factor model (Fama and French, 2015)
We prefer to use PI. If this is unavailable we use EBIT*-
Earnings before Tax and Extraordi-
pi* XINT+SPI+NOPI where we set SPI and NOPI to zero if
nary Items
missing
Income Tax tax Compustat item TXT
We prefer to use XIDO. If this is unavailable, we use XI+DO
Extraordinary Items and Discontinued where we set DO to zero if missing. The reason why we
xido*
Operations set missing DO to zero is because it is not available in
COMP.G FUNDQ
We prefer to use IB. If this is unavailable, we use NI-XIDO*.
Net Income ni* If this is unavailable, we prefer PI*-TXT-MII. If MII is un-
availble, it is set to zero
We prefer NI. If this is not available, we prefer NI*+XIDO*. If
Net Income Including Extraordinary
nix* XIDO* is unavailable, we set it to zero. If that is unavailable,
Items
we prefer NI*+XI+DO
Firm Income fi* We use NIX*+XINT
Dividends for Common Shareholds dvc Compustat Item DVC
Total Dividends div* We prefer DVT. If this is not available, we use DV
Income Before Extraordinary Items ni qtr* We use IBQ
Net Sales sale qtr* We use SALEQ
Cash Flow Statement
Capital Expenditures capx Compustat item CAPX
Capital Expenditures to Sales capex sale* We use CAPX / SALE*
We use OCF*-CAPX. Note that the free cash flow is com-
Free Cash Flow fcf* puted before financing activities and sale of assets is taken
into account

11
Name Abbreviation Construction
We use PRSTKC+PURTSHR Equity Buyback is mainly
Equity Buyback eqbb* PRSTKC in NA and PURTSHR in GLOBAL. Either of
PRSTKC or PURTSHR are allowed to be missing
Equity Issuance eqis* Compustat item SSTK
We use EQIS*-EQBB*. Either EQIS* or EQBB* are allowed
Equity Net Issuance eqnetis*
to be missing
Net Equity Payout eqpo* We use DIV*+EQBB*
Equity Net Payout eqnpo* We use DIV*-EQNETIS*
We prefer to use DLTIS-DLTR where we only require that
one of the items are non-missing. If this is unavailable, we
Net Long-Term Debt Issuance dltnetis*
use LTDCH. If this is unavailable we use the yearly change in
long-term book debt DLTT
We prefer DLCCH. If this is unavailable, we use the yearly
Net Short-Term Debt Issuance dstnetis*
change in short-term book debt DLC
We use DLTNETIS*+DSTNETIS* and only require one of
Net Debt Issuance dbnetis*
the items to be non-missing
We use EQNETIS*+DBNETIS* and require that both
Net Issuance netis*
EQNETIS* and DBNETIS* are non-missing
We prefer FINCF. If this is unavailable, we use NETIS*-
Financial Cash Flow fincf* DV+FIAO+TXBCOF. If FIAO or TXBCOF is missing, it
is set to zero
Balance Sheet - Assets
We prefer to use AT. If this is unavailable, then we use SEQ*
Total Assets at* + DLTT + LCT + LO + TXDITC. If LCT, LO, or TXDITC
are missing, then they are set to zero
We prefer ACT. If this is unavailable, we use
Current Assets ca*
RECT+INVT+CHE+ACO
Account Receivables rec Compustat item RECT
Cash and Short-Term Investment cash Compustat item CHE
Inventory inv Compustat item INVT
Non-Current Assets nca* We use AT* - CA*
Intangible Assets intan Compustat item INTAN
Investment and Advances ivao Compustat item IVAO
Property, Plans and Equipment Gross ppeg Compustat item PPEGT
Property, Plans and Equipment Net ppen Compustat item PPENT
Balance Sheet - Liabilities
Total Liabilities lt Compustat item LT
We prefer LCT. If this is unavailable, we use AP+ DLC+
Current Liabilities cl*
TXP+ LCO
Accounts Payable ap Compustat item AP
Short-Term Debt debtst Compustat item DLC
Income Tax Payable txp Compustat item TXP
Non-Current Liabilities ncl* We use LT-CL*
Long-Term Debt debtlt Compustat item DLTT
We prefer to use TXDITC. If this is unavailable, we use
Deferred Taxes and Investment Credit txditc*
TXDB+ ITCB
Balance Sheet - Financing
We prefer to use PSTKRV. If this is unavailable, we use
Preferred Stock pstk*
PSTKL. If this is unavilable, we use PSTK
We use DLTT+ DLC. Either DLTT or DLC are allowed to
Total Debt debt*
me missing
Net Debt netdebt* We use DEBT*- CHE where we set CHE to zero if missing
We prefer to use SEQ. If this is unavailable, we use
Shareholders Equity seq* CEQ+PSTK* where we set PSTK* to zero if missing. If this
is unavailable, we use AT- LT
We use SEQ*+TXDITC*-PSTK* where we set TXDITC*
Book Equity be*
and PSTK* to zero if missing
We prefer to use ICAPT+DLC-CHE where DLC and CHE
are set to zero if missing. If this is unavailable, we use
Book Enterprise Value bev*
SEQ*+NETDEBT*+ MIB where we set MIB to zero if miss-
ing. In the global data ICAPT is reduced by Treasury stock
Balance Sheet - Summary
Net Working Capital nwc* We use CA*-CL*
Current Operating Assets coa* We use CA*- CHE
Current Operating Liabilities col* We use CL*- DLC. If DLC is missing, it is set to zero
Current Operating Working Capital cowc* We use COA*-COL*
Non-Current Operating Assets ncoa* We use AT* - CA*- IVAO

12
 Name Abbreviation Construction
Non-Current Operating Liabilities ncol* We use LT-CL*- DLTT
Net Non-Current Operating Assets nncoa* We use NCOA*-NCOL*
Financial Assets fna* We use IVST+ IVAO. If either is missing, they are set to zero
We use DEBT*+PSTK*. If PSTK* is missing, it is set to
Financial Liabilities fnl*
zero
Net Financial Assets nfna* We use FNA*-FNL*
Operating Assets oa* We use COA*+NCOA*
Operating Liabilities ol* We use COL*+NCOL*
Net Operating Assets noa* We use OA*-OL*
Long-Term NOA lnoa* PPENT + INTAN + AO - LO + DP
We prefer to use CA* - INVT. If this is unavailable, we use
Liquid Current Assets caliq*
CHE + RECT
Property Plant and Equipment Less
ppeinv* PPEGT + INVT
Inventories
CHE + 0.75× COA* + 0.5(AT* - CA* - INTAN). If INTAN
Ortiz-Molina and Phillips Liquidity aliq*
is missing, we set it to zero
Market Based
We use the market equity for the stock we deem to the primary
security of the firm. Importantly, we do not align the market
Market Equity me value with the end of the fiscal period. Instead, we update the
market value on a monthly basis and align it with the most
recently available accounting characteristic
Market Enterprise Value mev* We use ME COMPANY + NETDEBT* × FX*
Market Assets mat* We use AT* × FX - BE* × FX + ME COMPANY
Accruals
We prefer NI*-OANCF. If that is unavailable, we use the
Operating Accruals oacc*
yearly change in COWC*+the yearly change in NNCOA*
Total Accruals tacc* We use OACC* + the yearly change in NFNA*
We prefer to use OANCF. If this is unavailable, we use NI*-
Operating Cash Flow ocf* OACC*. If this is unavailable, we use NI* + DP - WCAPT.
If WCAPT is missing, we use 0.
We use OANCFQ. If this is unavailable, then we use IBQ +
Quarterly Operating Cash Flow ocf qtr*
DPQ - WCAPTQ. If WCAPTQ is unavailable, we set it to
We prefer EBITDA*+XRD-OACC*. If XRD is unavailable,
Cash Based Operating Profitability cop*
we set it to zero
Other
Employees in Thousands emp Compustat item EMP

Table 6: Accounting Characteristics

Name Abbreviation Construction

Accounting Based Size Measures

Assets assets AT *t

Sales sales SALE*t

Book Equity book equity BE*t

Net Income net income N I*t

Enterprise Value enterprise value M EV *t

Growth - Percentage5
AT *t
Asset Growth 1yr at gr1 −1
AT *t−12

5
This refers to all variables with a suffix of “ gr1” or “ gr3”. The variables are percentage growth in the
accounting variables before the suffix. The number in the suffix refers to either 1 or 3 year growth. For all
variables, we only take the percentage growth if the denominator is above zero.

13
Name Abbreviation Construction
SALE *t
Sales Growth 1yr sale gr1 −1
SALE *t−12
CA*t
Current Asset Growth 1yr ca gr1 −1
CA*t−12
N CA*t
Non-Current Asset Growth 1yr nca gr1 −1
N CA*t−12
LTt
Total Liabilities Growth 1yr lt gr1 LTt−12
−1
CL*t
Current Liabilities Growth 1yr cl gr1 −1
CL*t−12
N CL*t
Non-Current Liabilities Growth 1yr ncl gr1 −1
N CL*t−12
BE *t
Book Equity Growth 1yr be gr1 −1
BE *t−12
P ST K *t
Preferred Stock Growth 1yr pstk gr1 −1
P ST K *t−12
DEBT *t
Total Debt Growth 1yr debt gr1 −1
DEBT *t−12
COGSt
Cost of Goods Sold Growth 1yr cogs gr1 COGSt−12
−1
XSGAt
Selling, General, and Administrative Expenses sga gr1 XSGAt−12
−1
Growth 1yr
OP EX *t
Operating Expenses Growth 1yr opex gr1 −1
OP EX *t−12
AT *t
Asset Growth 3yr at gr3 −1
AT *t−36
SALE *t
Sales Growth 3yr sale gr3 −1
SALE *t−36
CA*t
Current Asset Growth 3yr ca gr3 −1
CA*t−36
N CA*t
Non-Current Asset Growth 3yr nca gr3 −1
N CA*t−36
LTt
Total Liabilities Growth 3yr lt gr3 LTt−36
−1
CL*t
Current Liabilities Growth 3yr cl gr3 −1
CL*t−36
N CL*t
Non-Current Liabilities Growth 3yr ncl gr3 −1
N CL*t−36
BE *t
Book Equity Growth 3yr be gr3 −1
BE *t−36
P ST K *t
Preferred Stock Growth 3yr pstk gr3 −1
P ST K *t−36
DEBT *t
Total Debt Growth 3yr debt gr3 −1
DEBT *t−36
COGSt
Cost of Goods Sold Growth 3yr cogs gr3 COGSt−36
−1
XSGAt
Selling, General, and Administrative Expenses sga gr3 XSGAt−36
−1
Growth 3yr
OP EX *t
Operating Expenses Growth 3yr opex gr3 −1
OP EX *t−36

Growth - Changed Scaled by Total Assets

GP *t −GP *t−12
Gross Profit Change 1yr gp gr1a
AT *t
OCF *t −OCF *t−12
Operating Cash Flow Change 1yr ocf gr1a
AT *t
CASHt −CASHt−12
Cash and Short-Term Investments Change 1yr cash gr1a
AT *t
IN Vt −IN Vt−12
Inventory Change 1yr inv gr1a
AT *t
RECt −RECt−12
Receivables Change 1yr rec gr1a
AT *t

14
Name Abbreviation Construction
P P EGt −P P EGt−12
Property, Plans and Equiptment Gross ppeg gr1a
AT *t
Change 1yr
LT It −LT It−12
Investment and Advances Change 1yr lti gr1a
AT *t
IN T ANt −IN T ANt−12
Intangible Assets Change 1yr intan gr1a
AT *t
DEBT STt −DEBT STt−12
Short-Term Debt Change 1yr debtst gr1a
AT *t
APt −APt−12
Accounts Payable Change 1yr ap gr1a
AT *t
T XPt −T XPt−12
Income Tax Payable Change 1yr txp gr1a
AT *t
DEBT LTt −DEBT LTt−12
Long-Term Debt Change 1yr debtlt gr1a
AT *t
T XDIT C *t −T XDIT C *t−12
Deferred Taxes and Investment Credit Change txditc gr1a
AT *t
1yr
COA*t −COA*t−12
Current Operating Assets Change 1yr coa gr1a
AT *t
COL*t −COL*t−12
Current Operating Liabilities Change 1yr col gr1a
AT *t
COW C *t −COW C *t−12
Current Operating Working Capital Change cowc gr1a
AT *t
1yr
N COA*t −N COA*t−12
Non-Current Operating Assets Change 1yr ncoa gr1a
AT *t
N COL*t −N COL*t−12
Non-Current Operating Liabilities Change 1yr ncol gr1a
AT *t
N N COA*t −N N COA*t−12
Net Non-Current Operating Assets Change nncoa gr1a
AT *t
1yr
OA*t −OA*t−12
Operating Assets Change 1yr oa gr1a
AT *t
OL*t −OL*t−12
Operating Liabilities Change 1yr ol gr1a
AT *t
N OA*t −N OA*t−12
Net Operating Assets Change 1yr noa gr1a
AT *t
F N A*t −F N A*t−12
Financial Assets Change 1yr fna gr1a
AT *t
F N L*t −F N L*t−12
Financial Liabilities Change 1yr fnl gr1a
AT *t
N F N A*t −N F N A*t−12
Net Financial Assets Change 1yr nfna gr1a
AT *t
EBIT DA*t −EBIT DA*t−12
Operating Profit before Depreciation Change ebitda gr1a
AT *t
1yr
EBIT *t −EBIT *t−12
Operating Profit after Depreciation Change ebit gr1a
AT *t
1yr
OP E *t −OP E *t−12
Operating Earnings to Equity Change 1yr ope gr1a
AT *t
N I *t −N I *t−12
Net Income Change 1yr ni gr1a
AT *t
DPt −DPt−12
Depreciation and Amortization Change 1yr dp gr1a
AT *t
F CF *t −F CF *t−12
Free Cash Flow Change 1yr fcf gr1a
AT *t
N W C *t −N W C *t−12
Net Working Capital Change 1yr nwc gr1a
AT *t
N IX *t −N IX *t−12
Net Income Including Extraordinary Items nix gr1a
AT *t
Change 1yr
EQN ET IS *t −EQN ET IS *t−12
Equity Net Issuance Change 1yr eqnetis gr1a
AT *t

15
Name Abbreviation Construction
DLT N ET IS *t −DLT N ET IS *t−12
Net Long-Term Debt Issuance Change 1yr dltnetis gr1a
AT *t
DST N ET IS *t −DST N ET IS *t−12
Net Short-Term Debt Issuance Change 1yr dstnetis gr1a
AT *t
DBN ET IS *t −DBN ET IS *t−12
Net Debt Issuance Change 1yr dbnetis gr1a
AT *t
N ET IS *t −N ET IS *t−12
Net Issuance Change 1yr netis gr1a
AT *t
F IN CF *t −F IN CF *t−12
Financial Cash Flow Change 1yr fincf gr1a
AT *t
EQN P O *t −EQN P O *t−12
Equity Net Payout Change 1yr eqnpo gr1a
AT *t
T AXt −T AXt−12
Effective Tax Rate Change 1yr tax gr1a
AT *t
DIV *t −DIV *t−12
Dividend Payout Ratio Change 1yr div gr1a
AT *t
EQBB *t −EQBB *t−12
Equity Buyback Change 1yr eqbb gr1a
AT *t
EQIS *t −EQIS *t−12
Equity Issuance Change 1yr eqis gr1a
AT *t
EQP O *t −EQP O *t−12
Net Equity Payout Change 1yr eqpo gr1a
AT *t
CAP Xt −CAP Xt−12
Capital Expenditures Change 1yr capx gr1a
AT *t
GP *t −GP *t−36
Gross Profit Change 3yr gp gr3a
AT *t
OCF *t −OCF *t−36
Operating Cash Flow Change 3yr ocf gr3a
AT *t
CASHt −CASHt−36
Cash and Short-Term Investments Change 3yr cash gr3a
AT *t
IN Vt −IN Vt−36
Inventory Change 3yr inv gr3a
AT *t
RECt −RECt−36
Receivables Change 3yr rec gr3a
AT *t
P P EGt −P P EGt−36
Property, Plans and Equipment Gross Change ppeg gr3a
AT *t
3yr
LT It −LT It−36
Investment and Advances Change 3yr lti gr3a
AT *t
IN T ANt −IN T ANt−36
Intangible Assets Change 3yr intan gr3a
AT *t
DEBT STt −DEBT STt−36
Short-Term Debt Change 3yr debst gr3a
AT *t
APt −APt−36
Accounts Payable Change 3yr ap gr3a
AT *t
T XPt −T XPt−36
Income Tax Payable Change 3yr txp gr3a
AT *t
DEBT LTt −DEBT LTt−36
Long-Term Debt Change 3yr debtlt gr3a
AT *t
T XDIT C *t −T XDIT C *t−36
Deferred Taxes and Investment Credit Change txditc gr3a
AT *t
3yr
COA*t −COA*t−36
Current Operating Assets Change 3yr coa gr3a
AT *t
COL*t −COL*t−36
Current Operating Liabilities Change 3yr col gr3a
AT *t
COW C *t −COW C *t−36
Current Operating Working Capital Change cowc gr3a
AT *t
3yr
N COA*t −N COA*t−36
Non-Current Operating Assets Change 3yr ncoa gr3a
AT *t
N N COA*t −N N COA*t−36
Net Non-Current Operating Assets Change nncoa gr3a
AT *t
3yr
OA*t −OA*t−36
Operating Assets Change 3yr oa gr3a
AT *t

16
Name Abbreviation Construction
OL*t −OL*t−36
Operating Liabilities Change 3yr ol gr3a
AT *t
N OA*t −N OA*t−36
Net Operating Assets Change 3yr noa gr3a
AT *t
F N A*t −F N A*t−36
Financial Assets Change 3yr fna gr3a
AT *t
F N L*t −F N L*t−36
Financial Liabilities Change 3yr fnl gr3a
AT *t
N F N A*t −N F N A*t−36
Net Financial Assets Change 3yr nfna gr3a
AT *t
EBIT DA*t −EBIT DA*t−36
Operating Profit before Depreciation Change ebitda gr3a
AT *t
3yr
EBIT *t −EBIT *t−36
Operating Profit after Depreciation Change ebit gr3a
AT *t
3yr
OP E *t −OP E *t−36
Operating Earnings to Equity Change 3yr ope gr3a
AT *t
N I *t −N I *t−36
Net Income Change 3yr ni gr3a
AT *t
DPt −DPt−36
Depreciation and Amortization Change 3yr dp gr3a
AT *t
F CF *t −F CF *t−36
Free Cash Flow Change 3yr fcf gr3a
AT *t
N W C *t −N W C *t−36
Net Working Capital Change 3yr nwc gr3a
AT *t
IN Vt −IN Vt−36
Inventory Change 1yr inv gr3a
AT *t
N COL*t −N COL*t−36
Non-Current Operating Liabilities Change 3yr ncol gr3a
AT *t
N IX *t −N IX *t−36
Net Income Including Extraordinary Items nix gr3a
AT *t
Change 3yr
EQN ET IS *t −EQN ET IS *t−36
Equity Net Issuance Change 3yr eqnetis gr3a
AT *t
DLT N ET IS *t −DLT N ET IS *t−36
Net Long-Term Debt Issuance Change 3yr dltnetis gr3a
AT *t
DST N ET IS *t −DST N ET IS *t−36
Net Short-Term Debt Issuance Change 3yr dstnetis gr3a
AT *t
DBN ET IS *t −DBN ET IS *t−36
Net Debt Issuance Change 3yr dbnetis gr3a
AT *t
N ET IS *t −N ET IS *t−36
Net Issuance Change 3yr netis gr3a
AT *t
F IN CF *t −F IN CF *t−36
Financial Cash Flow Change 3yr fincf gr3a
AT *t
N W C *t −N W C *t−36
Net Working Capital Change 3yr nwc gr3a
AT *t
EQN P O *t −EQN P O *t−36
Equity Net Payout Change 3yr eqnpo gr3a AT t
T AXt −T AXt−36
Effective Tax Rate Change 3yr tax gr3a AT t
DIV *t −DIV *t−36
Dividend Payout Ratio Change 3yr div gr3a AT t
EQBB *t −EQBB *t−36
Equity Buyback Change 3yr eqbb gr3a AT t
EQIS *t −EQIS *t−36
Equity Issuance Change 3yr eqis gr3a AT t
EQP O *t −EQP O *t−36
Net Equity Payout Change 3yr eqpo gr3a AT t
CAP Xt −CAP Xt−36
Capital Expenditures Change 3yr capx gr3a AT t

Investment
CAP Xt
Capital Expenditures scaled by Assets capx at
AT *t

17
Name Abbreviation Construction
XRDt
R&D scaled by Assets rd at
AT *t

Non-Recurring Items
SP It
Special Items scaled by Assets spi at
AT *t
XIDO *t
Extraordinary Items and Discontinued Opera- xido at
AT *t
tions scaled by Assets
SP It +XIDO *t
Non-Recurring Items scaled by Assets nri at
AT *t

Profit Margins
GP *t
Gross Profit Margin gp sale
SALE *t
EBIT DA*t
Operating Profit Margin before Depreciation ebitda sale
SALE *t
EBIT *t
Operating Profit Margin after Depreciation ebit sale
SALE *t
P I *t
Pretax Profit Margin pi sale
SALE *t
N I *t
Net Profit Margin before XI ni sale
SALE *t
N IX *t
Net Profit Margin nix sale
SALE *t
F CF *t
Free Cash Flow Margin fcf sale
SALE *t
OCF *t
Operating Cash Flow Margin ocf sale
SALE *t

Return on Assets
GP *t
Gross Profit scaled by Assets gp at
AT *t
EBIT DA*t
Operating Profit before Depreciation scaled by ebitda at
AT *t
Assets
EBIT *t
Operating Profit after Depreciation scaled by ebit at
AT *t
Assets
F I *t
Firm Income scaled by Assets fi at
AT *t
COP *t
Cash Based Operating Profitability scaled by cop at
AT *t
Assets

Return on Book Equity

OP E *t
Operating Profit to Equity scaled by BE ope be
BE *t
N I *t
Net Income scaled by BE ni be
BE *t
N IX *t
Net Income Including Extraordinary Items nix be
BE *t
scaled by BE
OCF *t
Operating Cash Flow scaled by BE ocf be
BE *t
F CF *t
Free Cash Flow scaled by BE fcf be
BE *t

Return on Invested Capital

GP *t
Gross Profit scaled by BEV gp bev
BEV *t
EBIT DA*t
Operating Profit before Depreciation scaled by ebitda bev
BEV *t
BEV

18
Name Abbreviation Construction
EBIT *t
Operating Profit after Depreciation scaled by ebit bev
BEV *t
BEV
F I *t
Firm Income scaled by BEV fi bev
BEV *t
COP *t
Cash Based Operating Profitability scaled by cop bev
BEV *t
BEV

Return on Physical Capital

GP *t
Gross Profit scaled by PPEN gp ppen P P ENt

EBIT DA*t
Operating Profit before Depreciation scaled by ebitda ppen P P ENt
PPEN
F CF *t
Free Cash Flow scaled by PPEN fcf ppen P P ENt

Issuance
F IN CF *t
Financial Cash Flow scaled by Assets fincf at
AT *t
N ET IS *t
Net Issuance scaled by Assets netis at
AT *t
EQN ET IS *t
Equity Net Issuance scaled by Assets eqnetis at
AT *t
EQIS *t
Equity Issuance scaled by Assets eqis at
AT *t
DBN ET IS *t
Net Debt Issuance scaled by Assets dbnetis at
AT *t
DLT N ET IS *t
Net Long-Term Debt Issuance scaled by Assets dltnetis at
AT *t
DST N ET IS *t
Net Short-Term Debt Issuance scaled by As- dstnetis at
AT *t
sets

Equity Payout
EQN P O *t
Equity Net Payout scaled by Assets eqnpo at
AT *t
EQBB *t
Net Equity Payout scaled by Assets eqbb at
AT *t
DIV *t
Total Dividends scaled by Assets div at
AT *t

Accruals
OACC *t
Operating Accruals oaccruals at
AT *t
OACC *t
Percent Operating Accruals oaccruals ni
|N IX *t |
T ACC *t
Total Accruals taccruals at
AT *t
T ACC *t
Percent Total Accruals taccruals ni
|N IX *t |
N OA*t
Net Operating Asset to Total Assets noa at
AT *t

Capitalization/Leverage Ratios
BE *t
Common Equity scaled by BEV be bev
BEV *t
DEBT *t
Total Debt scaled by BEV debt bev
BEV *t
CASHt
Cash and Short-Term Investments scaled by cash bev
BEV *t
BEV
P ST K *t
Preferred Stock scaled by BEV pstk bev
BEV *t

19
Name Abbreviation Construction
DEBT LTt
Long-Term Debt scaled by BEV debtlt bev
BEV *t
DEBT STt
Short-Term Debt scaled by BEV debtst bev
BEV *t
DEBT *t
Total Debt scaled by MEV debt mev
M EV *t
P ST K *t
Preferred Stock scaled by MEV pstk mev
M EV *t
DEBT LTt
Long-Term Debt scaled by MEV debtlt mev
M EV *t
DEBT STt
Short-Term Debt scaled by MEV debtst mev
M EV *t

Financial Soundness Ratios

IN Tt
Interest scaled by Total Debt int debt
DEBT *t
IN Tt
Interest scaled by Long-Term Debt int debtlt DEBT LTt

EBIT DA*t
Operating Profit before Depreciation scaled by ebitda debt
DEBT *t
Total Debt
EBIT DA*t
Profit before D&A scaled by Current Liabili- profit cl
CL*t
ties
OCF *t
Operating Cash Flow scaled by Current Lia- ocf cl
CL*t
bilities
OCF *t
Operating Cash Flow scaled by Total Debt ocf debt
DEBT *t
CASHt
Cash Balance scaled by Total Liabilities cash lt LTt

IN Vt
Inventory scaled by Current Assets inv act ACTt

RECt
Receivables scaled by Current Assets rec act ACTt

DEBT STt
Short-Term Debt scaled by Total Debt debtst debt
DEBT *t
CL*t
Current Liabilities scaled by Total Liabilities cl lt LTt

DEBT LTt
Long-Term Debt scaled by Total Debt debtlt debt
DEBT *t
OP EX *t
Operating Leverage opex at
AT *t
F CF *t
Free Cash Flow scaled by Operating Cash fcf ocf
OCF *t
Flow
LTt
Total Liabilities scaled by Total Tangible As- lt ppen P P ENt
sets
DEBT LTt
Long-Term Debt to Book Equity debtlt be
BE *t
N W C *t
Working Capital scaled by Assets nwc at
AT *t

Solvency Ratios
DEBT *t
Debt-to-Assets debt at
AT *t
DEBT *t
Debt to Shareholders’ Equity Ratio debt be
BE *t
EBIT *t
Interest Coverage Ratio ebit int IN Tt

Liquidity Ratios
IN Vt +IN Vt−12
2
Days Inventory Outstanding inv days COGSt
× 365

20
Name Abbreviation Construction
RECt +RECt−12
2
Days Sales Outstanding rec days × 365
SALE t *
APt +APt−12
2
Days Accounts Payable Outstanding ap days COGSt
× 365

Cash Conversion Cycle cash conversion IN V DAY St + REC DAY St − AP DAY St

CASHt
Cash Ratio cash cl
CL*t
CALIQ*t
Quick Ratio caliq cl
CL*t
CA*t
Current Ratio ca cl
CL*t

Activity/Efficiency Ratios
COGSt
Inventory Turnover inv turnover (IN Vt +IN Vt−12 )/2

SALE *t
Asset Turnover at turnover
(AT *t +AT *t−12 )/2
SALE *t
Receivables Turnover rec turnover (RECt +RECt−12 )/2
COGSt +IN Vt −IN Vt−12
Account Payables Turnover ap turnover (APt +APt−12 )/2

Miscellaneous
XADt
Advertising scaled by Sales adv sale
SALE *t
XLRt
Labor Expense scaled by Sales staff sale
SALE *t
SALE *t
Sales scaled by BEV sale bev
BEV *t
XRDt
R&D scaled by Sales rd sale
SALE *t
SALE *t
Sales scaled by Total Stockholders’ Equity sale be
BE *t
DV Ct
Dividend Payout Ratio div ni
N I *t
SALE *t
Sales scaled by Working Capital sale nwc
N W C *t
T AXt
Effective Tax Rate tax pi
P I *t

Balance Sheet Fundamental to Market Equity

BE *t
Book Equity scaled by Market Equity be me M Et

AT *t
Total Assets scaled by Market Equity at me M Et

CASHt
Cash and Short-Term Investments scaled by cash me M Et
Market Equity

Income Fundamentals to Market Equity

GP *t
Gross Profit scaled by ME gp me M Et

EBIT DA*t
Operating Profit before Depreciation scaled by ebitda me M Et
ME
EBIT *t
Operating Profit after Depreciation scaled by ebit me M Et
ME
OP E *t
Operating Earnings to Equity scaled by ME ope me M Et

N I *t
Net Income scaled by ME ni me M Et

SALE *t
Sales scaled by ME sale me M Et

21
Name Abbreviation Construction
OCF *t
Operating Cash Flow scaled by ME ocf me M Et

F CF *t
Free Cash Flow scaled by ME fcf me M Et

N IX *t
Net Income Including Extraordinary Items nix me M Et
scaled by ME
COP *t
Cash Based Operating Profitability scaled by cop me M Et
ME
XRDt
R&D scaled by ME rd me M Et

Balance Sheet Fundamentals to Market Enterprise Value

BE *t
Book Equity scaled by MEV be mev
M EV *t
AT *t
Total Assets scaled by MEV at mev
M EV *t
CASHt
Cash and Short-Term Investments scaled by cash mev
M EV *t
MEV
BEV *t
Book Enterprise Value scaled by MEV bev mev
M EV *t
P P ENt
Property, Plans and Equipment Net scaled by ppen mev
M EV *t
MEV

Equity Payout/Issuance to Market Equity

DIV *t
Total Dividends scaled by ME div me M Et

EQBB *t
Equity Buyback scaled by ME eqbb me M Et

EQIS *t
Equity Issuance scaled by ME eqis me M Et

EQP O *t
Net Equity Payout scaled by ME eqpo me M Et

EQN P O *t
Equity Net Payout scaled by ME eqnpo me M Et

EQN ET IS *t
Equity Net Issuance scaled by ME eqnetis me M Et

Debt Issuance to Market Enterprise Value

DLT N ET IS *t
Net Long-Term Debt Issuance scaled by MEV dltnetis mev
M EV *t
DST N ET IS *t
Net Short-Term Debt Issuance scaled by MEV dstnetis mev
M EV *t
DBN ET IS *t
Net Debt Issuance scaled by MEV dbnetis mev
M EV *t

Firm Payout/Issuance to Market Enterprise Value

N ET IS *t
Net Issuance scaled by MEV netis mev
M EV *t

Income Fundamentals to Market Enterprise Value

GP *t
Gross Profit scaled by MEV gp mev
M EV *t
EBIT DA*t
Operating Profit before Depreciation scaled by ebitda mev
M EV *t
MEV
EBIT *t
Operating Profit after Depreciation scaled by ebit mev
M EV *t
MEV
SALE *t
Sales scaled by MEV sale mev
M EV *t
OCF *t
Operating Cash Flow scaled by MEV ocf mev
M EV *t

22
Name Abbreviation Construction
F CF *t
Free Cash Flow scaled by MEV fcf mev
M EV *t
COP *t
Cash Based Operating Profitability scaled by cop mev
M EV *t
MEV
F IN CF *t
Financial Cash Flow Change scaled by MEV fincf mev
M EV *t

New Variables not in HXZ

QT R*t
 
NI
Net Income to Sales Quarterly Volatility niq saleq std σ8Q
SALE QT R*t
N I *t
Net Income scaled by Employees ni emp EM Pt

SALE *t
Sales scaled by Employees sale emp EM Pt

N I *t
Net Income scaled by Assets ni at
AT *t
OCF *t
Operating Cash Flow scaled by Assets ocf at
AT *t

Operating Cash Flow to Assets 1 yr Change ocf at chg1 OCF ATt − OCF ATt−12

σ16Q N I QT R*t
 
Quarterly ROE Volatility roeq be std
BE *t
 
ROE Volatility roe be std σ60M N I *t
BE *t
GP *t GP *t−60
Gross Product to Assets 5 yr Change gpoa ch5 −
AT *t AT *t−60
N I *t N I *t−60
ROE 5 yr Change roe ch5 −
BE *t BE *t−60
N I *t N I *t−60
ROA 5 yr Change roa ch5 −
AT *t AT *t−60
OCF *t OCF *t−60
Operating Cash Flow to Assets 5 yr Change cfoa ch5 −
AT *t AT *t−60
GP *t GP *t−60
Gross Product to Sales 5 yr Change gmar ch5 −
SALE *t SALE *t−60

New Variables from HXZ

CASHt
Cash and Short Term Investments scaled by cash at
AT *t
Assets

Number of Consecutive Earnings Increases ni inc8q Count number of earnings increases over past 8 quarters
P P EIN V *t −P P EIN V *t−12
Change in Property, Plant and Equipment ppeinv gr1a
AT *t−12
Less Inventories scaled by lagged Assets
LN OA*t −LN OA*t−12
Change in Long-Term NOA scaled by average lnoa gr1a
AT *t −AT *t−12
Assets
CAP Xt
CAPX 1 year growth capx gr1 CAP Xt−12
−1
CAP Xt
CAPX 2 year growth capx gr2 CAP Xt−24
−1
CAP Xt
CAPX 3 year growth capx gr3 CAP Xt−36
−1
IV STt −IV STt−12
Change in Short-Term Investments scaled by sti gr1a
AT *t
Assets
N I QT R*t
Quarterly Income scaled by BE niq be
BE *t−3

Change in Quarterly Income scaled by BE niq be chg1 N IQ BEt − N IQ BEt−12

N I QT R*t
Quarterly Income scaled by AT niq at
AT *t−3

Change in Quarterly Income scaled by AT niq at chg1 N IQ ATt − N IQ ATt−12

23
Name Abbreviation Construction
SALE QT R*t
Quarterly Sales Growth saleq gr1 −1
SALE QT R*t−12
P4
n=0 (1−.2×n)(XRDt−12∗n )
R&D Capital-to-Assets rd5 at
AT *t

Age age Age of the firms in months

Change Sales minus Change Inventory dsale dinv CHG T O EXP (SALE*t ) − CHG T O EXP (IN Vt )

Change Sales minus Change Receivables dsale drec CHG T O EXP (SALE*t ) − CHG T O EXP (RECt )

Change Gross Profit minus Change Sales dgp dsale CHG T O EXP (GP *t ) − CHG T O EXP (SALE*t )

Change Sales minus Change SG&A dsale dsga CHG T O EXP (SALE*t ) − CHG T O EXP (XSGAt )

Earnings Surprise saleq su SU R(SALE QT R*)

Revenue Surprrise niq su SU R(N I QT R*)

DEBT *t
Total Debt scaled by ME debt me M Et

N ET DEBT *t
Net Debt scaled by ME netdebt me M Et

Abnormal Corporate Investment capex abn CAP X SALE t * −1

* * *
(CAP X SALE t−12 +CAP X SALE t−24 +CAP X SALE t−36 )/3

IN Vt
Inventory Change 1 yr inv gr1 IN Vt−12
−1
BE *t −BE *t−12
Book Equity Change 1 yr scaled by Assets be gr1a
AT *t
OP *t
Ball Operating Profit to Assets op at
AT *t
P I *t
Earnings before Tax and Extraordinary Items pi nix
N IX *t
to Net Income Including Extraordinary Items
OP *t
Ball Operating Profit scaled by lagged Assets op atl1
AT *t−12
OP E *t
Operating Profit scaled by lagged Book Equity ope bel1
BE *t−12
GP *t
Gross Profit scaled by lagged Assets gp atl1
AT *t−12
COP *t
Cash Based Operating Profitability scaled by cop atl1
AT *t−12
lagged Assets
AT *t
Book Leverage at be
BE *t
OCF QT R*t
 
Operating Cash Flow to Sales Quarterly ocfq saleq std σ16Q )
SALE QT R*t
Volatility
ALIQ*t
Liquidity scaled by lagged Assets aliq at
AT *t−12
ALIQ*t
Liquidity scaled by lagged Market Assets aliq mat
M AT *t−12
CASHt +0.715×RECt +0.547×IN Vt +0.535×P P EGt
Tangibility tangibility
AT *t

Equity Duration eq dur Outlined in detail here

Piotroski F-Score f score Outlined in detail here

Ohlson O-Score o score Outlined in detail here

Altman Z-Score z score Outlined in detail here

Kaplan-Zingales Index kz index Outlined in detail here

Intrinsic Value intrinsic value Outlined in detail here

24
 Name Abbreviation Construction
IN T RIN SIC V ALU E *t
Intrinsic value-to-market ival me M Et

SALE EM Pt
Sales scaled by Employees Growth 1 yr sale emp gr1 SALE EM Pt−12
−1
EM Pt −EM Pt−12
Employee Growth 1 yr emp gr1 0.5×EM Pt +0.5×EM Pt−12

σ60M N I *t /AT *t−12
Earnings Variability earnings variability  
σ60M OCF *t /AT *t−12

cov((N I */AT *)t ,(N I */AT *)t−12 )

1 yr lagged Net Income to Assets ni ar1
var((N I */AT *)t−12 )

Net Income Idiosyncratic Volatility ni ivol Outlined in detail here

7 Market Based Characteristics

Datasets

• CRSP.MSF

• CRSP.DSF

• COMP.SECD

• COMP.G SECD

• COMP.SECM

• COMP.SECURITY

• COMP.G SECURITY

Market Variables
A suffix of ’*’ indicates that we have altered or renamed the original item.
Table 7: Market Variables
Name Abbreviation Construction
CRSP Variables6
Share Adjustment Factor adjfct* We use CFACSHR
Shares shares* We use SHROUT/1000 so shares outstanding are in millions.
Price prc* We use |PRC|
Local Price prc local* We use PRC*
We use ASKHI. If PRC* or AKSHI are negative, then
Highest Daily Price prc high
PRC HIGH is set to missing
We use BIDLO. If PRC* or BIDLO are negative, then
Lowest Daily Price prc low
PRC LOW is set to missing
We use PRC*×SHARES* so market equity is quoted in mil-
Market Equity me*
lion USD.
Company Market Equity me company* We sum ME* grouped by PERMCO and date
Trading Volume tvol* We use VOL
Dollar Volume dolvol* We use TVOL*×PRC*
We use RET. In case of delisting, we calculate as
Return RET*
(1+RET)*(1+DLRET)-1

6
lag is a lag function where lag(x) is the value of x from the previous time period

25
Name Abbreviation Construction
We use RET. In case of delisting, we calculate as
Local Return ret local*
(1+RET)*(1+DLRET)-1
We use RET*-T30RET/21. If T30RET is unavailable, we
Excess Return ret exc* use RF. If the return is a monthly return rather than a daily
return, the T30RET is divided by 1 rather than 21.
Excess Return t+1 ret exc lead1m* Excess return (ret exc*) in month t+1
Time Since Most Recent Return ret lag dif* We automatically set this to 1
Cumulative Return ri* This is the cumulative return estimated from RET*
We use
Monthly Dividend div tot*
(RET -RETX)×lag(PRC*)×(CFACSHR/lag(CFACSHR))
Compustat Variables
Share Adjustment Factor adjfct* We use AJEXDI
Shares shares* We use CSHOC/1000000
Price prc* We use PRC LOCAL*×FX
Local Price prc local* We use PRCCD
Market Equity me* We use PRC*×SHARES*
Company Market Equity me company* We use ME*
Trading Volume tvol* We use CSHTRD
Dollar Volume dolvol* We use TVOL*×PRC*
Cumulative Return - Local ri local* We use PRC LOCAL*× TRFD/AJEXDI
We use RI LOCAL*/lag(RI LOCAL*) - 1. In case of delisting,
Local Return ret local* we calculate as (RI LOCAL*/lag(RI LOCAL*) * (1+dlret7 )
- 1)
Cumulative Return ri* RI LOCAL* × FX*
We use RI*/lag(RI*) - 1. In case of delisting, we calculate as
Return RET*
(RI*/lag(RI*) * (1+dlret) -1)
We use RET*-T30RET/21. If T30RET is unavailable, we
Excess Return ret exc* use RF. If the return is a monthly return rather than a daily
return, the T30RET is divided by 1 rather than 21.
Excess Return t+1 ret exc lead1m* Excess return (ret exc*) in month t+1
We estimate the number of days since the previous return. If
Time Since Most Recent Return ret lag dif* the returns are monthly rather than daily, then the time is in
months
Monthly Dividend div tot* We use DIV × FX*. If DIV is missing, we set it to zero
We use DIVD × FX*. If DIVD is unavailable, we set it to
Cash Dividend div cash*
zero
We use DIVSP × FX*. If DIVSP is unavailable, we set it to
Special Cash Dividend div spc*
zero
Bid-Ask Average Dummy bidask* When PRCSTD = 4 then 1, otherwise 0
Asset Pricing Factors
Excess Market Return mktrf* Country specific market return
Country specific factor following Fama and French (1993) and
High Minus Low hml* using breakpoints from non-micro cap stocks within the coun-
try
Average of small portfolios minus average of large portfolios
Small Minus Big ala Fama-French smb ff*
from hml*
Country specific factor following Hou, Xue and Zhang (2015)
and using breakpoints from non-micro cap stocks within the
Return on Equity roe* country. We use double sorts on return on equity and size
rather than triple sorts with investment, due to the limited
number of stocks in some international markets.
Country specific factor following Hou, Xue and Zhang (2015)
and using breakpoints from non-micro cap stocks within the
Investment inv* country. We use double sorts on investment and size rather
than triple sorts with return on equity, due to the limited
number of stocks in some international markets
Average of small portfolios minus average of large portfolios
Small Minus Big ala Hou et al smb hxz*
from roe* and inv*
Market Volatility for Each Stock mktvol zd* σzD (M KT RF *t ) 8

7
dlret is set to -0.3 when dlsrni is ’02’ or ’03’ and set to 0 otherwise
8
Must have enough non-missing values of stock to be estimated

26
 Table 8: Market Characteristics

Name Abbreviation Construction

Market Based Size Measures

Market Equity market equity M E*t

Total Dividend Paid to Market Equity

DIV T OT *t ×SHARES *t
Dividend to Price - 1 Month div1m me
M E *t
P2
n=0 DIV T OT *t−n ×SHARES *t−n
Dividend to Price - 3 Months div3m me
M E *t
P5
n=0 DIV T OT *t−n ×SHARES *t−n
Dividend to Price - 6 Months div6m me
M E *t
P11
n=0 DIV T OT *t−n ×SHARES *t−n
Dividend to Price - 12 Months div12m me
M E *t

Special Dividend Paid to Market Equity

DIV SP C *t ×SHARES *t
Special Dividend to Price - 1 Month divspc1m me
M E *t
P11
n=0 DIV SP C *t−n ×SHARES *t−n
Special Dividend to Price - 12 Month divspc12m me
M E *t

Change in Shares Outstanding

SHARES *t ×ADJF CT *t
Change in Shares - 1 Month chcsho 1m −1
SHARES *t−1 ×ADJF CT *t−1
SHARES *t ×ADJF CT *t
Change in Shares - 3 Month chcsho 3m −1
SHARES *t−3 ×ADJF CT *t−3
SHARES *t ×ADJF CT *t
Change in Shares - 6 Month chcsho 6m −1
SHARES *t−6 ×ADJF CT *t−6
SHARES *t ×ADJF CT *t
Change in Shares - 12 Month chcsho 12m −1
SHARES *t−12 ×ADJF CT *t−12

Net Equity Payout

   
RI *t M E *t
Net Equity Payout - 1 Month eqnpo 1m log − log
RI *t−1 M E *t−1
   
RI *t M E *t
Net Equity Payout - 3 Month eqnpo 3m log − log
RI *t−3 M E *t−3
   
RI *t M E *t
Net Equity Payout - 6 Month eqnpo 6m log − log
RI *t−6 M E *t−6
   
RI *t M E *t
Net Equity Payout - 12 Month eqnpo 12m log − log
RI *t−12 M E *t−12

Momentum/Reversal
RI *t
Short Term Reversal ret 1 0 −1
RI *t−1
RI *t
Momentum 0-2 Months ret 2 0 −1
RI *t−2
RI *t
Momentum 0-3 Months ret 3 0 −1
RI *t−3
RI *t−1
Momentum 1-3 Months ret 3 1 −1
RI *t−3
RI *t
Momentum 0-6 Months ret 6 0 −1
RI *t−6
RI *t−1
Momentum 1-6 Months ret 6 1 −1
RI *t−6
RI *t
Momentum 0-9 Months ret 9 0 −1
RI *t−9
RI *t−1
Momentum 1-9 Months ret 9 1 −1
RI *t−9
RI *t
Momentum 0-12 Months ret 12 0 −1
RI *t−12

27
 Name Abbreviation Construction
RI *t−1
Momentum 1-12 Months ret 12 1 −1
RI *t−12
RI *t−7
Momentum 7-12 Months ret 12 7 −1
RI *t−12
RI *t−1
Momentum 1-18 Months ret 18 1 −1
RI *t−18
RI *t−1
Momentum 1-24 Months ret 24 1 −1
RI *t−24
RI *t−12
Momentum 12-24 Months ret 24 12 −1
RI *t−24
RI *t−1
Momentum 1-36 Months ret 36 1 −1
RI *t−36
RI *t−12
Momentum 12-36 Months ret 36 12 −1
RI *t−36
RI *t−1
Momentum 1-48 Months ret 48 1 −1
RI *t−48
RI *t−12
Momentum 12-48 Months ret 48 12 −1
RI *t−48
RI *t−1
Momentum 1-60 Months ret 60 1 −1
RI *t−60
RI *t−12
Momentum 12-60 Months ret 60 12 −1
RI *t−60
RI *t−36
Momentum 36-60 Months ret 60 36 −1
RI *t−60

Seasonality

1 Year Annual Seasonality seas 1 1an Return in month t-12

2 - 5 Year Annual Seasonality seas 2 5an Average return over annual lags from year t-2 to t-5

6 - 10 Year Annual Seasonality seas 6 10an Average return over annual lags from year t-6 to t-10

11 - 15 Year Annual Seasonality seas 11 15an Average return over annual lags from year t-11 to t-15

16 - 20 Year Annual Seasonality seas 16 20an Average return over annual lags from year t-16 to t-20)

1 Year Non-Annual Seasonality seas 1 1na Average return from month t-1 to t-11

2 - 5 Year Non-Annual Seasonality seas 2 5na Average return over non-annual lags from year t-2 to t-5

6 - 10 Year Non-Annual Seasonality seas 6 10na Average return over non-annual lags from year t-6 to t-10

11 - 15 Year Non-Annual Seasonality seas 11 15na Average return over non-annual lags from year t-11 to t-15

16 - 20 Year Non-Annual Seasonality seas 16 20na Average return over non-annual lags from year t-16 to t-20

Combined Accounting and Market Based Characteristics

Let et be defined as described here

COV AR 60M (RET *t ,M KT RF *t )
60 Month CAPM Beta beta 60m
V ARC 60M (M KT RF *t )
1
O SCOREtr01 + RET 12 1r01 t +
Performance Based Mispricing mispricing perf9 4
GP ATtr01 + N IQ ATtr01

1
CHCSHO 12Mtr01 + EQN P O 12Mtr01 +
6
Management Based Mispricing mispricing mgmt OACCRU ALS ATtr01 + N OA ATtr01 +
AT GR1r01 + P P EIN V GR1Ar01


t t

9
A rank characteristic has the value of that characteristics rank with respect to other companies’ same
characteristic of the same month and country scaled [0, 1]. This is identified with a “r01” superscript.

28
 Name Abbreviation Construction
Q6
Residual Momentum - 6 Month resff3 6 1 −1 + n=1 1 + et−n
Q12
Residual Momentum - 12 Month resff3 12 1 −1 + n=1 1 + et−n

Daily Market Data10

Let ϵt be defined as described here

Return Volatility rvol zd σzD (RET EXC*t )

Maximum Return rmax1 zd M AX1 zD(RET *t )

1 P5
Mean Maximum Return rmax5 zd 5 n=1 Xn , Xn ∈ M AX5 zD(RET *)

Return Skewness rskew zd SKEW zD(RET EXC*t )

P RC ADJ *t
Price-to-High prc highprc zd
M AX1 zD(P RC ADJ *t )
 
|RET *t |
Amihud (2002) Measure ami zd ∗ 1000000
DOLV OL*t
zD

CAPM Beta beta zd Described in detail here

CAPM Idiosyncratic Vol. ivol capm zd Described in detail here

CAPM Skewness iskew capm zd Described in detail here

(ϵt ×M KT RF DMt2 )zD
Coskewness coskew zd11 q
(ϵ2 2
t )zD ×(M KT RF DMt )zD

Fama and French Idiosyncratic Vol. ivol ff3 zd Described in detail here

Fama and French Skewness iskew ff3 zd Described in detail here

Hou, Xue and Zhang Idiosyncratic Vol. ivol hxz4 zd Described in detail here

Hou, Xue and Zhang Skewness iskew hxz4 zd Described in detail here

Dimson Beta beta dimson zd Created as described in Dimson (1979)

Downside Beta betadown zd Described in detail here

Number of days with zero trades over period. In case of equal

Zero Trades zero trades zd number of zero trading days, turnover zd will decide on the
rank following Liu (2006)
 
T V OL*t
Turnover turnover zd
SHARES
 *t ∗1000000 zD 
σzD (T V OL*t /SHARES *t )∗1000000
Turnover Volatility turnover var zd T U RN OV ER zDt

Dollar Volume dolvol zd DOLV OL*tzD

σzD (DOLV OL*t )
Dollar Volume Volatility dolvol var zd DOLV OL zDt

The correlation between RET EXC*3 l = RET EXC*t +

Correlation to Market corr zd RET EXC*t−1 + RET EXC*t−2 and M KT EXC 3l =
M KT RF *t + M KT RF *t−1 + M KT RF *t−2
CORR 1260dt ×RV OL 252dt
Betting Against Beta betabab 1260d
M KT V OL 252d*t

10
Many of the variables in this section are estimated using rolling windows of data, and the variables are
estimated using a variety of window lengths: 21, 126, 252 and 1260 days. In this section, I refer to the
number of days as m as a proxy for any of the possible window lengths.
11
M KT RF DMt = M KT RF *t − M KT RF *tzD

29
 Name Abbreviation Construction
RM AX5 21dt
Max Return to Volatility rmax5 rvol 21d RV OL 252dt

High-low bid ask estimator created using code from Corwin

21 Day Bid-Ask High-Low bidaskhl 21d
and Schultz (2012)

High-low return volatility estimator created using code from

21 Day Return Volatility High-Low rvolhl 21d
Corwin and Schultz (2012)

Quality Minus Junk


ZV ZV (GP ATt ) + ZV (N I BEt )+
Quality Minus Junk - Profit qmj prof ZV (N I ATt ) + ZV (OCF ATt ) + ZV (GP SALE*t )+

ZV (OACCRU ALS ATt )


ZV ZV (GP OA CH5t ) + ZV (ROE CH5t )
Quality Minus Junk - Growth qmj growth +ZV (ROA CH5t ) + ZV (CF OA CH5t )+

ZV (GM AR CH5t )


ZV ZV (BET ABAB 1260dt ) + ZV (DEBT ATt )
Quality Minus Junk - Safety qmj safety 
+ZV (O SCOREt ) + ZV (Z SCOREt ) + ZV ( EV OLt )
QM J P ROFt +QM J GROW T Ht +QM J SAF ET Yt
Quality Minus Junk qmj 3

8 Detailed Characteristic Construction

This section includes detailed descriptions how we built characteristics that don’t easily fit
into the Accounting Characteristics or Market Characteristics tables.

• Equity Duration

– Define the following variables:

∗ horizon: number of months used to estimate helper variables
∗ r: constant used as assumed discount rate
∗ roe mean: constant used as the average ROE value
∗ roe ar1: constant used as the expected growth rate of ROE
∗ g mean: constant used as the average sales growth rate
∗ g ar1: constant used as the expected growth rate of sales
– Create initial variables:
N I*t
roe0 =
BE*t−12
SALE*t
g0 = −1
SALE*t−12
be0 = BE*t

30
 ∗ If the number of non-missing observations is less than or equal to 12 or the
variables’ respective denominators are less than or equal to 1 roe0t and g0t
are set to missing.
– Forecast cash distributions

roe c = roe mean × (1 − roe ar1)

g c = g mean × (1 − g ar1)
horizon
X
roet = roe c + roe ar1 × roet−i
i=1
horizon
X
gt = g c + g ar1 × gt−i
i=1
horizon
X
bet = bet−i × (1 + gt )
i=1
horizon
X
cdt = bet × ( roet − gt )
i=1

12
– Create duration helper variables

1+r
ed constant = horizon +
r
horizon
X cdt
ed cw wt = ed cd wi−1 + i ×
i=1
(1 + r)i
horizon
X cdt
ed cdt = ed cdi−1 +
i=1
(1 + r)i

– Characteristic:
ed ed wt × F Xt M E COM P AN Yt − ed cdt × F Xt
eq durt = +ed constant×
M E COM P AN Yt M E COM P AN Yt

• Piotroski F-Score
12
ed cw w, ed cd and ed err are equal to 0 at i = 1. ed cw w and ed cd recusrively build upon themselves
over the length of the horiozon, so ed cw wi−1 , for example, would be the previous iteration of ed cw w

31
 – Create helper variables:

N I*t
f roat =
AT *t−12
OCF *t
f croat =
AT *t−12
f droat = f roat − f roat−12
f acct = f croat − f roat
DLT Tt DLT Tt−12
f lev = −
AT *t AT *t−12
CA*t CA*t−12
f liqt = −
CL*t CL*t−12
f eqist = EQIS*t
GP *t GP *t−12
f gmt = −
SALE*t SALE*t−12
SALE*t SALE*t−12
f aturnt = −
AT *t−12 AT *t−24
∗ For all variables except f acc, f aturn f eqis, if the count of available
observations is less than or equal to 12, then the variable is set to missing.
If f aturn has less than or equal to 24 non-missing observations, it is set
to missing. If a variable has AT *t or AT *t−12 as an input and AT *t ≤ 0 or
AT *t−12 ≤ 0, then it is set to missing. If CL*t ≤ 0 or CL*t−12 ≤ 0 then
f liqt is set to missing. If SALE*t ≤ 0 or SALE*t−12 ≤ 0 then f gmt is
set to missing.
– Characteristic13

f scoret = f roa>0,t + f croa>0,t + f droa>0,t + f acc>0,t +

f lev<0,t + f liq>0,t + f eqis=0,t + f gm>0,t + f aturn>0,t

• Ohlson O-Score
13
A subscript of > 0, ex: V ARt>0,t , is a dummy for if the variable is greater than zero, and it is defined
similarly for V ARt<0,t or any other specification. Otherwise, not included as an input, Also, if any variables
other than f eqist are missing, then f scoret is set to missing.

32
 – Create helper variables:

o latt = AT *t−1
DEBT *t
o levt =
AT *t
CA*t − CL*t
o wct =
AT *t
N IX*t
o roet =
AT *t
CL*t
o caclt =
CA*t
P I*t + DPt
o f f ot =
LTt
o neg eqt = 1 if LTt > AT *t , otherwise 0
o neg earn = 1 if N IX*t < 0 and N IX*t−12 < 0
N IX*t − N IX*t−12
o nicht =
|N IX*t | + |N IX*t−12 |

∗ If AT *t ≤ 0, then o latt , o levt , o wct , and o roet are set to missing. If

CA*t ≤ 0 then o caclt is set to missing. If LTT ≤ 0 then o f f ot is set to
missing. If LTt or AT *t are missing, then o neg eqt is set to missing. If there
are less than or equal to 12 observations or either of N IX*t and N IX*t−12
are missing, then o nicht and o neg earnt are set to missing.
– Characteristic:

o scoret = − 1.37 − 0.407 × o latt + 6.03 × o levt + 1.43 × o wct +

0.076 × o caclt − 1.72 × o neg eqt − 2.37 × o roet −
1.83 × o f f ot + 0.285 × o neg earnt − 0.52 × o nicht

• Altman Z-Score

– Create helper variables:

CA*t − CL*t
z wct =
AT *t
REt
z ret =
AT *t
EBIT DA*t
z ebt =
AT *t
SALE*t
z sat =
AT *t
M E F ISCALt
z met =
LTt

33
 ∗ If AT *t ≤ 0 then any variable including AT *t , then it is set to missing. If
LTt ≤ 0, then z met is set to missing.
– Characteristic:

z scoret = 1.2 × z wct + 1.4 × z ret + 3.3 × z ebt + 0.6 × z met + 1.0 × z sat

• Kaplan-Zingales Index
– Create helper variables:
N I*t + DPt
kz cft =
P P EN Tt−12
AT *t + M E F ISCALt − BE*t
kz qt =
AT *t
DEBT *t
kz dbt =
DEBT *t + SEQ*t
DIV *t
kz dvt =
P P EN Tt−12
CHEt
kz cst =
P P EN Tt−12
∗ If the number of non-missing observations is less than or equal to 12, then
kz cft , kz dvt and kz cst are set to zero. If P P EN Tt−12 ≤ 0 then kz cft ,
kz dvt and kz cst are set to missing. If AT *t ≤ 0 then kz qt is set to
missing. If (DEBT *t + SEQ*t ) = 0 then kz dbt is set to missing.
– Characteristic:

kz index = −1.002× kz cft +0.283× kz qt +3.139× kz dbt −39.368× kz dvt −1.315× kz cst

• Intrinsic Value from Frankel and Lee

– Define r as a constant assumed discount rate
– Create helper variables:
DIV *t
iv pot =
N IX*t
N IX*t
iv roet =
(BE*t + BE*t−12 )/2
iv be1t = (1 + (1 − iv pot ) × iv roet ) × BE*t )

∗ If N IX*t ≤ 0 then
DIV *t
iv pot =
AT *t × 0.06
∗ If the number of non-missing observations is less than or equal to 12 or
(BE*t + BE*t−12 ) ≤ 0 then iv roet is set to missing.

34
 – Characteristics:
iv roet − r iv roet − r
intrinsic valuet = BE*t + × BE*t + × iv be1t
1+r (1 + r) × r

∗ If intrinsic valuet ≤ 0 then it is set to missing.

• Net Income Idiosyncratic Volatility

14
– Define the following variable :

N I*t
ni att =
AT *t

– A rolling regression of the following form is run for each company, with the time
series split up into n groups:

ni att = β0 + β1 ni att−12 + ut

where edft = the error degrees of freedom of regression and rmset = root mean
square error of the regression.
– Characteristic:
s
rmse2t × edft
ni ivolt =
edft + 1

• Beta, Idiosyncratic Volatility and Skewness of Asset Pricing Factor Regressions

– This section describes the construction of beta zd for the CAPM model, and the
idiosyncratic volatility and skewness characteristics, which are estimated using
three different factor models:
∗ CAPM (capm):

RET EXC*t = β0 + β1 M KT RF *t + ϵt

∗ Fama-French 3 Factor Model (ff3):

RET EXC*t = β0 + β1 M KT RF *t + β2 HM L*t + β3 SM B F F *t + et

∗ Hou, Xue and Zhang 4 Factor Model (hxz4):

RET EXC*t = β0 + β1 M KT RF *t + β2 SM B HXZ*t + β3 ROE*t + β4 IN V *t + µt

14
If AT *t ≤ 0, then ni att is set to missing

35
 15
– Characteristics :

beta zd = β1 from the CAPM model

ivol capm zdt = σzD (ϵt )
ivol f f 3 zdt = σzD (et )
ivol hxz4 zdt = σzD (µt )
iskew capm zdt = SKEW zD(ϵt )
iskew f f 3 zdt = SKEW zD(et )
iskew hxz4 zdt = SKEW zD(σt )

• Downside Beta

– Define the following regression model run over z days:

RET EXC*t = β0 + β1 M KT RF *t + ϵt

However, we restrict the data to when M KT RF * is negative.

– Characteristic:
∗ betadown zd = β1

9 FX Conversion Rate Construction

This section outlines how we create a daily dataset, beginning 01/01/1950 to now, of X
currency - USD exchange rate using COMPUSTAT. This is run in the macro compustat f x()
in the project macros.sas file.

• We use COMP.EXRT DLY, which has daily conversion rates from GBP to other cur-
rencies ’X’.

• Every day available, we estimate the exchange rate f xt as

U SDGBP,t
f xt =
XGBP,t

where XGBP,t is the exchange rate of GBP to currency X on day t.

• In case there are gaps in information, we assume the exchange rate of the last obser-
vation until a new observation is available.

• f xt is quoted as Xt
U SDt
, so to go from X to USD, do Xt × f xt
15
z indicates over how many days the model is run.

36
10 Factor Details and Citations
Table 9: Factor and Cluster Details
Variable Orig. Orig.
Description Name Citation Sample Sign Signif.
Accruals
Change in current operating work- cowc gr1a Richardson, Sloan, Soliman, and 1962-2001 -1 1
ing capital Tuna (2005)
Operating accruals oaccruals at Sloan (1996) 1962-1991 -1 1
Percent operating accruals oaccruals ni Hafzalla, Lundholm, and Matthew 1989-2008 -1 1
Van Winkle (2011)
Years 16-20 lagged returns, nonan- seas 16 20na Heston and Sadka (2008) 1965-2002 1 1
nual
Total accruals taccruals at Richardson et al. (2005) 1962-2001 -1 1
Percent total accruals taccruals ni Hafzalla et al. (2011) 1989-2008 -1 1

Debt Issuance
Abnormal corporate investment capex abn Titman, Wei, and Xie (2004) 1973-1996 -1 1
Growth in book debt (3 years) debt gr3 Lyandres, Sun, and Zhang (2008) 1970-2005 -1 1
Change in financial liabilities fnl gr1a Richardson et al. (2005) 1962-2001 -1 1
Change in noncurrent operating li- ncol gr1a Richardson et al. (2005) 1962-2001 -1 0
abilities
Change in net financial assets nfna gr1a Richardson et al. (2005) 1962-2001 1 1
Earnings persistence ni ar1 Francis, LaFond, Olsson, and 1975-2001 1 0
Schipper (2004)
Net operating assets noa at Hirshleifer, Hou, Teoh, and Zhang 1964-2002 -1 1
(2004)

Investment
Liquidity of book assets aliq at Ortiz-Molina and Phillips (2014) 1984-2006 -1 0
Asset Growth at gr1 Cooper, Gulen, and Schill (2008) 1968-2003 -1 1
Change in common equity be gr1a Richardson et al. (2005) 1962-2001 -1 1
CAPEX growth (1 year) capx gr1 Xie (2001) 1971-1992 -1 0
CAPEX growth (2 years) capx gr2 Anderson and Garcia-Feijoo (2006) 1976-1998 -1 1
CAPEX growth (3 years) capx gr3 Anderson and Garcia-Feijoo (2006) 1976-1998 -1 1
Change in current operating assets coa gr1a Richardson et al. (2005) 1962-2001 -1 1
Change in current operating liabil- col gr1a Richardson et al. (2005) 1962-2001 -1 1
ities
Hiring rate emp gr1 Belo, Lin, and Bazdresch (2014) 1965-2010 -1 1
Inventory growth inv gr1 Belo and Lin (2012) 1965-2009 -1 1
Inventory change inv gr1a J. K. Thomas and Zhang (2002) 1970-1997 -1 1
Change in long-term net operating lnoa gr1a Fairfield, Whisenant, and Yohn 1964-1993 -1 1
assets (2003)
Mispricing factor: Management mispricing mgmtStambaugh and Yuan (2017) 1967-2013 1 1
Change in noncurrent operating as- ncoa gr1a Richardson et al. (2005) 1962-2001 -1 1
sets
Change in net noncurrent operating nncoa gr1a Richardson et al. (2005) 1962-2001 -1 1
assets
Change in net operating assets noa gr1a Hirshleifer et al. (2004) 1964-2002 -1 1
Change PPE and Inventory ppeinv gr1a Lyandres et al. (2008) 1970-2005 -1 1
Long-term reversal ret 60 12 De Bondt and Thaler (1985) 1926-1982 -1 1

37
Sales Growth (1 year) sale gr1 Lakonishok, Shleifer, and Vishny 1968-1989 -1 1
(1994)
Sales Growth (3 years) sale gr3 Lakonishok et al. (1994) 1968-1989 -1 1
Sales growth (1 quarter) saleq gr1 1967-2016 -1 0
Years 2-5 lagged returns, nonannual seas 2 5na Heston and Sadka (2008) 1965-2002 -1 1

Low Leverage
Firm age age Jiang, Lee, and Zhang (2005) 1965-2001 -1 1
Liquidity of market assets aliq mat Ortiz-Molina and Phillips (2014) 1984-2006 -1 0
Book leverage at be Fama and French (1992) 1963-1990 -1 0
The high-low bid-ask spread bidaskhl 21d Corwin and Schultz (2012) 1927-2006 1 1
Cash-to-assets cash at Palazzo (2012) 1972-2009 1 0
Net debt-to-price netdebt me Penman, Richardson, and Tuna 1962-2001 -1 1
(2007)
Earnings volatility ni ivol Francis et al. (2004) 1975-2001 1 0
R&D-to-sales rd sale Chan, Lakonishok, and Sougiannis 1975-1995 1 0
(2001)
R&D capital-to-book assets rd5 at Li (2011) 1952-2004 1 0
Asset tangibility tangibility Hahn and Lee (2009) 1973-2001 1 0
Altman Z-score z score Dichev (1998) 1981-1995 1 1

Low Risk
Market Beta beta 60m Fama and MacBeth (1973) 1935-1968 -1 1
Dimson beta beta dimson 21d Dimson (1979) 1955-1974 -1 0
Frazzini-Pedersen market beta betabab 1260d Frazzini and Pedersen (2014) 1926-2012 -1 1
Downside beta betadown 252d Ang, Chen, and Xing (2006) 1963-2001 -1 1
Earnings variability earnings variability
Francis et al. (2004) 1975-2001 -1 0
Idiosyncratic volatility from the ivol capm 21d 1967-2016 -1 0
CAPM (21 days)
Idiosyncratic volatility from the ivol capm 252d Ali, Hwang, and Trombley (2003) 1976-1997 -1 1
CAPM (252 days)
Idiosyncratic volatility from the ivol ff3 21d Ang, Hodrick, Xing, and Zhang 1963-2000 -1 1
Fama-French 3-factor model (2006)
Idiosyncratic volatility from the q- ivol hxz4 21d 1967-2016 -1 0
factor model
Cash flow volatility ocfq saleq std Huang (2009) 1980-2004 -1 1
Maximum daily return rmax1 21d Bali, Cakici, and Whitelaw (2011) 1962-2005 -1 1
Highest 5 days of return rmax5 21d Bali, Brown, and Tang (2017) 1993-2012 -1 1
Return volatility rvol 21d Ang, Hodrick, et al. (2006) 1963-2000 -1 1
Years 6-10 lagged returns, nonan- seas 6 10na Heston and Sadka (2008) 1965-2002 -1 1
nual
Share turnover turnover 126d Datar, Naik, and Radcliffe (1998) 1963-1991 -1 1
Number of zero trades with zero trades 21d Liu (2006) 1963-2003 1 0
turnover as tiebreaker (1 month)
Number of zero trades with zero trades 126d Liu (2006) 1963-2003 1 1
turnover as tiebreaker (6 months)
Number of zero trades with zero trades 252d Liu (2006) 1963-2003 1 1
turnover as tiebreaker (12 months)

Momentum
Current price to high price over last prc highprc 252dGeorge and Hwang (2004) 1963-2001 1 1
year
Residual momentum t-6 to t-1 resff3 6 1 Blitz, Huij, and Martens (2011) 1930-2009 1 1
Residual momentum t-12 to t-1 resff3 12 1 Blitz et al. (2011) 1930-2009 1 1

38
Price momentum t-3 to t-1 ret 3 1 Jegadeesh and Titman (1993) 1965-1989 1 1
Price momentum t-6 to t-1 ret 6 1 Jegadeesh and Titman (1993) 1965-1989 1 1
Price momentum t-9 to t-1 ret 9 1 Jegadeesh and Titman (1993) 1965-1989 1 1
Price momentum t-12 to t-1 ret 12 1 Jegadeesh and Titman (1993) 1965-1989 1 1
Year 1-lagged return, nonannual seas 1 1na Heston and Sadka (2008) 1965-2002 1 1

Profit Growth
Change sales minus change Inven- dsale dinv Abarbanell and Bushee (1998) 1974-1988 1 1
tory
Change sales minus change receiv- dsale drec Abarbanell and Bushee (1998) 1974-1988 -1 0
ables
Change sales minus change SG&A dsale dsga Abarbanell and Bushee (1998) 1974-1988 1 0
Change in quarterly return on as- niq at chg1 1972-2016 1 0
sets
Change in quarterly return on eq- niq be chg1 1967-2016 1 0
uity
Standardized earnings surprise niq su Foster, Olsen, and Shevlin (1984) 1974-1981 1 1
Change in operating cash flow to as- ocf at chg1 Bouchaud, Krueger, Landier, and 1990-2015 1 1
sets Thesmar (2019)
Price momentum t-12 to t-7 ret 12 7 Novy-Marx (2012) 1925-2010 1 1
Labor force efficiency sale emp gr1 Abarbanell and Bushee (1998) 1974-1988 1 0
Standardized Revenue surprise saleq su Jegadeesh and Livnat (2006) 1987-2003 1 1
Year 1-lagged return, annual seas 1 1an Heston and Sadka (2008) 1965-2002 1 1
Tax expense surprise tax gr1a J. Thomas and Zhang (2011) 1977-2006 1 1

Profitability
Coefficient of variation for dollar dolvol var 126d Chordia, Subrahmanyam, and An- 1966-1995 -1 1
trading volume shuman (2001)
Return on net operating assets ebit bev Soliman (2008) 1984-2002 1 1
Profit margin ebit sale Soliman (2008) 1984-2002 1 1
Pitroski F-score f score Piotroski (2000) 1976-1996 1 1
Return on equity ni be Haugen and Baker (1996) 1979-1993 1 1
Quarterly return on equity niq be Hou, Xue, and Zhang (2015) 1972-2012 1 1
Ohlson O-score o score Dichev (1998) 1981-1995 -1 1
Operating cash flow to assets ocf at Bouchaud et al. (2019) 1990-2015 1 1
Operating profits-to-book equity ope be Fama and French (2015) 1963-2013 1 1
Operating profits-to-lagged book ope bel1 1967-2016 1 0
equity
Coefficient of variation for share turnover var 126dChordia et al. (2001) 1966-1995 -1 1
turnover

Quality
Capital turnover at turnover Haugen and Baker (1996) 1979-1993 1 0
Cash-based operating profits-to- cop at 1967-2016 1 0
book assets
Cash-based operating profits-to- cop atl1 Ball, Gerakos, Linnainmaa, and 1963-2014 1 1
lagged book assets Nikolaev (2016)
Change gross margin minus change dgp dsale Abarbanell and Bushee (1998) 1974-1988 1 0
sales
Gross profits-to-assets gp at Novy-Marx (2013) 1963-2010 1 1
Gross profits-to-lagged assets gp atl1 1967-2016 1 0
Mispricing factor: Performance mispricing perf Stambaugh and Yuan (2017) 1967-2013 1 1
Number of consecutive quarters ni inc8q Barth, Elliott, and Finn (1999) 1982-1992 1 0
with earnings increases

39
Quarterly return on assets niq at Balakrishnan, Bartov, and Faurel 1976-2005 1 1
(2010)
Operating profits-to-book assets op at 1963-2013 1 1
Operating profits-to-lagged book op atl1 Ball et al. (2016) 1963-2014 1 1
assets
Operating leverage opex at Novy-Marx (2011) 1963-2008 1 1
Quality minus Junk: Composite qmj C. S. Asness, Frazzini, and Peder- 1957-2016 1 1
sen (2019)
Quality minus Junk: Growth qmj growth C. S. Asness et al. (2019) 1957-2016 1 1
Quality minus Junk: Profitability qmj prof C. S. Asness et al. (2019) 1957-2016 1 1
Quality minus Junk: Safety qmj safety C. S. Asness et al. (2019) 1957-2016 1 1
Assets turnover sale bev Soliman (2008) 1984-2002 1 1

Seasonality
Market correlation corr 1260d C. Asness, Frazzini, Gormsen, and 1925-2015 -1 1
Pedersen (2020)
Coskewness coskew 21d Harvey and Siddique (2000) 1963-1993 -1 1
Net debt issuance dbnetis at Bradshaw, Richardson, and Sloan 1971-2000 -1 1
(2006)
Kaplan-Zingales index kz index Lamont, Polk, and Saaá-Requejo 1968-1995 1 1
(2001)
Change in long-term investments lti gr1a Richardson et al. (2005) 1962-2001 -1 1
Taxable income-to-book income pi nix Lev and Nissim (2004) 1973-2000 1 1
Years 2-5 lagged returns, annual seas 2 5an Heston and Sadka (2008) 1965-2002 1 1
Years 6-10 lagged returns, annual seas 6 10an Heston and Sadka (2008) 1965-2002 1 1
Years 11-15 lagged returns, annual seas 11 15an Heston and Sadka (2008) 1965-2002 1 1
Years 11-15 lagged returns, nonan- seas 11 15na Heston and Sadka (2008) 1965-2002 -1 0
nual
Years 16-20 lagged returns, annual seas 16 20an Heston and Sadka (2008) 1965-2002 -1 1
Change in short-term investments sti gr1a Richardson et al. (2005) 1962-2001 1 0

Size
Amihud Measure ami 126d Amihud (2002) 1964-1997 1 1
Dollar trading volume dolvol 126d Brennan, Chordia, and Subrah- 1966-1995 -1 1
manyam (1998)
Market Equity market equity Banz (1981) 1926-1975 -1 1
Price per share prc Miller and Scholes (1982) 1940-1978 -1 1
R&D-to-market rd me Chan et al. (2001) 1975-1995 1 1

Short-Term Reversal
Idiosyncratic skewness from the iskew capm 21d 1967-2016 -1 0
CAPM
Idiosyncratic skewness from the iskew ff3 21d Bali, Engle, and Murray (2016) 1925-2021 -1 1
Fama-French 3-factor model
Idiosyncratic skewness from the q- iskew hxz4 21d 1967-2016 -1 0
factor model
Short-term reversal ret 1 0 Jegadeesh (1990) 1929-1982 -1 1
Highest 5 days of return scaled by rmax5 rvol 21d C. Asness et al. (2020) 1925-2015 -1 1
volatility
Total skewness rskew 21d Bali et al. (2016) 1925-2021 -1 1

Value
Assets-to-market at me Fama and French (1992) 1963-1990 1 0

40
Book-to-market equity be me Rosenberg, Reid, and Lanstein 1973-1984 1 1
(1985)
Book-to-market enterprise value bev mev Penman et al. (2007) 1962-2001 1 1
Net stock issues chcsho 12m Pontiff and Woodgate (2008) 1970-2003 -1 1
Debt-to-market debt me Bhandari (1988) 1948-1979 1 1
Dividend yield div12m me Litzenberger and Ramaswamy 1940-1980 1 1
(1979)
Ebitda-to-market enterprise value ebitda mev Loughran and Wellman (2011) 1963-2009 1 1
Equity duration eq dur Dechow, Sloan, and Soliman (2004) 1962-1998 -1 1
Net equity issuance eqnetis at Bradshaw et al. (2006) 1971-2000 -1 1
Equity net payout eqnpo 12m Daniel and Titman (2006) 1968-2003 1 1
Net payout yield eqnpo me Boudoukh, Michaely, Richardson, 1984-2003 1 1
and Roberts (2007)
Payout yield eqpo me Boudoukh et al. (2007) 1984-2003 1 1
Free cash flow-to-price fcf me Lakonishok et al. (1994) 1963-1990 1 1
Intrinsic value-to-market ival me Frankel and Lee (1998) 1975-1993 1 0
Net total issuance netis at Bradshaw et al. (2006) 1971-2000 -1 1
Earnings-to-price ni me Basu (1983) 1963-1979 1 1
Operating cash flow-to-market ocf me Desai, Rajgopal, and Venkatacha- 1973-1997 1 1
lam (2004)
Sales-to-market sale me Barbee Jr, Mukherji, and Raines 1979-1991 1 1
(1996)

Other Factors
Assets assets
Sales sales
Book Equity book equity
Net Income net income
Enterprise Value enterprise value
Current Asset Growth 1yr ca gr1
Non-Current Asset Growth 1yr nca gr1
Total Liabilities Growth 1yr lt gr1
Current Liabilities Growth 1yr cl gr1
Non-Current Liabilities Growth 1yr ncl gr1
Book Equity Growth 1yr be gr1
Preferred Stock Growth 1 yr pstk gr1
Total Debt Growth 1yr debt gr1
Cost of Goods Sold Growth 1yr cogs gr1
Selling, General, and Administra- sga gr1
tive Expenses Growth 1yr
Operating Expenses Growth 1yr opex gr1
Asset Growth 3yr at gr3
Current Asset Growth 3yr ca gr3
Non-Current Asset Growth 3yr nca gr3
Total Liabilities Growth 3yr lt gr3
Current Liabilities Growth 3yr cl gr3
Non-Current Liabilities Growth 3yr ncl gr3
Book Equity Growth 3yr be gr3
Preferred Stock Growth 3yr pstk gr3
Cost of Goods Sold Growth 3yr cogs gr3
Selling, General, and Administra- sga gr3
tive Expenses Growth 3yr
Operating Expenses Growth 3yr opex gr3
Gross Profit Change 1yr gp gr1a

41
Operating Cash Flow Change 1yr ocf gr1a
Cash and Short-Term Investments cash gr1a
Change 1yr
Receivables Change 1yr rec gr1a
Property, Plans and Equipment ppeg gr1a
Gross Change 1yr
Intangible Assets Change 1yr intan gr1a
Short-Term Debt Change 1yr debtst gr1a
Accounts Payable Change 1yr ap gr1a
Income Tax Payable Change 1yr txp gr1a
Long-Term Debt Change 1yr debtlt gr1a
Deferred Taxes and Investment txditc gr1a
Credit Change 1yr
Non-Current Operating Liabilities ncol gr1a
Change 1yr
Operating Assets Change 1yr oa gr1a
Operating Liabilities Change 1yr ol gr1a
Financial Assets Change 1yr fna gr1a
Operating Profit before Deprecia- ebitda gr1a
tion Change 1yr
Operating Profit after Depreciation ebit gr1a
Change 1yr
Operating Earnings to Equity ope gr1a
Change 1yr
Net Income Change 1yr ni gr1a
Depreciation and Amortization dp gr1a
Change 1yr
Free Cash Flow Change 1yr fcf gr1a
Net Working Capital Change 1yr nwc gr1a
Net Income Including Extraordi- nix gr1a
nary Items Change 1yr
Equity Net Issuance Change 1yr eqnetis gr1a
Net Long-Term Debt Issuance dltnetis gr1a
Change 1yr
Net Short-Term Debt Issuance dstnetis gr1a
Change 1yr
Net Debt Issuance Change 1yr dbnetis gr1a
Net Issuance Change 1yr netis gr1a
Financial Cash Flow Change 1yr fincf gr1a
Equity Net Payout Change 1yr eqnpo gr1a
Dividend Payout Ratio Change 1yr div gr1a
Equity Buyback Change 1yr eqbb gr1a
Equity Issuance Change 1yr eqis gr1a
Net Equity Payout Change 1yr eqpo gr1a
Capital Expenditures Change 1yr capx gr1a
Gross Profit Change 3yr gp gr3a
Operating Cash Flow Change 3yr ocf gr3a
Cash and Short-Term Investments cash gr3a
Change 3yr
Inventory Change 3yr inv gr3a
Receivables Change 3yr rec gr3a
Property, Plans and Equipment ppeg gr3a
Gross Change 3yr

42
Investment and Advances Change lti gr3a
3yr
Intangible Assets Change 3yr intan gr3a
Short-Term Debt Change 3yr debtst gr3a
Accounts Payable Change 3yr ap gr3a
Income Tax Payable Change 3yr txp gr3a
Long-Term Debt Change 3yr debtlt gr3a
Deferred Taxes and Investment txditc gr3a
Credit Change 3yr
Current Operating Assets Change coa gr3a
3yr
Current Operating Liabilities col gr3a
Change 3yr
Current Operating Working Capi- cowc gr3a
tal Change 3yr
Non-Current Operating Assets ncoa gr3a
Change 3yr
Net Non-Current Operating Assets nncoa gr3a
Change 3yr
Operating Assets Change 3yr oa gr3a
Operating Liabilities Change 3yr ol gr3a
Net Operating Assets Change 3yr noa gr3a
Financial Assets Change 3yr fna gr3a
Financial Liabilities Change 3yr fnl gr3a
Net Financial Assets Change 3yr nfna gr3a
Operating Profit before Deprecia- ebitda gr3a
tion Change 3yr
Operating Profit after Depreciation ebit gr3a
Change 3yr
Operating Earnings to Equity ope gr3a
Change 3yr
Net Income Change 3yr ni gr3a
Depreciation and Amortization dp gr3a
Change 3yr
Free Cash Flow Change 3yr fcf gr3a
Net Working Capital Change 3yr nwc gr3a
Inventory Change 1yr inv gr3a
Non-Current Operating Liabilities ncol gr3a
Change 3yr
Net Income Including Extraordi- nix gr3a
nary Items Change 3yr
Equity Net Issuance Change 3yr eqnetis gr3a
Net Long-Term Debt Issuance dltnetis gr3a
Change 3yr
Net Short-Term Debt Issuance dstnetis gr3a
Change 3yr
Net Debt Issuance Change 3yr dbnetis gr3a
Net Issuance Change 3yr netis gr3a
Financial Cash Flow Change 3yr fincf gr3a
Net Working Capital Change 3yr nwc gr3a
Equity Net Payout Change 3yr eqnpo gr3a
Effective Tax Rate Change 3yr tax gr3a
Dividend Payout Ratio Change 3yr div gr3a
Equity Buyback Change 3yr eqbb gr3a

43
Equity Issuance Change 3yr eqis gr3a
Net Equity Payout Change 3yr eqpo gr3a
Capital Expenditures Change 3yr capx gr3a
Capital Expenditures scaled by As- capx at
sets
R&D scaled by Assets rd at
Special Items scaled by Assets spi at
Extraordinary Items and Discontin- xido at
ued Operations scaled by Assets
Non-Recurring Items scaled by As- nri at
sets
Gross Profit Margin gp sale
Operating Profit Margin before De- ebitda sale
preciation
Pretax Profit Margin pi sale
Net Profit Margin before extraordi- ni sale
nary income
Net Profit Margin nix sale
Free Cash Flow Margin fcf sale
Operating Cash Flow Margin ocf sale
Operating Profit before Deprecia- ebitda at
tion scaled by Assets
Operating Profit after Depreciation ebit at
scaled by Assets
Firm Income scaled by Assets fi at
Net Income Including Extraordi- nix be
nary Items scaled by BE
Operating Cash Flow scaled by BE ocf be
Free Cash Flow scaled by BE fcf be
Gross Profit scaled by BEV gp bev
Operating Profit before Deprecia- ebitda bev
tion scaled by BEV
Firm Income scaled by BEV fi bev
Cash Based Operating Profitability cop bev
scaled by BEV
Gross Profit scaled by PPEN gp ppen
Operating Profit before Deprecia- ebitda ppen
tion scaled by PPEN
Free Cash Flow scaled by PPEN fcf ppen
Financial Cash Flow scaled by As- fincf at
sets
Equity Issuance scaled by Assets eqis at
Net Long-Term Debt Issuance dltnetis at
scaled by Assets
Net Short-Term Debt Issuance dstnetis at
scaled by Assets
Equity Net Payout scaled by Assets eqnpo at
Net Equity Payout scaled by Assets eqbb at
Total Dividends scaled by Assets div at
Common Equity scaled by BEV be bev
Total Debt scaled by BEV debt bev
Cash and Short-Term Investments cash bev
scaled by BEV
Preferred Stock scaled by BEV pstk bev

44
Long-Term Debt scaled by BEV debtlt bev
Short-Term Debt scaled by BEV debtst bev
Total Debt scaled by MEV debt mev
Preferred Stock scaled by MEV pstk mev
Long-Term Debt scaled by MEV debtlt mev
Short-Term Debt scaled by MEV debtst mev
Interest scaled by Total Debt int debt
Interest scaled by Long-Term Debt int debtlt
Operating Profit before Deprecia- ebitda debt
tion scaled by Total Debt
Profit before D&A scaled by Cur- profit cl
rent Liabilities
Operating Cash Flow scaled by ocf cl
Current Liabilities
Operating Cash Flow scaled by To- ocf debt
tal Debt
Cash Balance scaled by Total Lia- cash lt
bilities
Inventory scaled by Current Assets inv act
Receivables scaled by Current As- rec act
sets
Short-Term Debt scaled by Total debtst debt
Debt
Current Liabilities scaled by Total cl lt
Liabilities
Long-Term Debt scaled by Total debtlt debt
Debt
Free Cash Flow scaled by Operating fcf ocf
Cash Flow
Total Liabilities scaled by Total lt ppen
Tangible Assets
Long-Term Debt to Book Equity debtlt be
Working Capital scaled by Assets nwc at
Debt-to-Assets debt at
Debt to Shareholders’ Equity Ratio debt be
Interest Coverage Ratio ebit int
Days Inventory Outstanding inv days
Days Sales Outstanding rec days
Days Accounts Payable Outstand- ap days
ing
Cash Conversion Cycle cash conversion
Cash Ratio cash cl
Quick Ratio caliq cl
Current Ratio ca cl
Inventory Turnover inv turnover
Receivables Turnover rec turnover
Account Payables Turnover ap turnover
Advertising scaled by Sales adv sale
Labor Expense scaled by Sales staff sale
Sales scaled by Total Stockholders’ sale be
Equity
Dividend Payout Ratio div ni
Sales scaled by Working Capital sale nwc
Effective Tax Rate tax pi

45
Intrinsic Value intrinsic value
Cash and Short-Term Investments cash me
scaled by Market Equity
Gross Profit scaled by ME gp me
Operating Profit before Deprecia- ebitda me
tion scaled by ME
Operating Profit after Depreciation ebit me
scaled by ME
Operating Earnings to Equity ope me
scaled by ME
Net Income Including Extraordi- nix me
nary Items scaled by ME
Cash Based Operating Profitability cop me
scaled by ME
Book Equity scaled by MEV be mev
Total Assets scaled by MEV at mev
Cash and Short-Term Investments cash mev
scaled by MEV
Property, Plans and Equipment ppen mev
Net scaled by MEV
Total Dividends scaled by ME div me
Equity Buyback scaled by ME eqbb me
Equity Issuance scaled by ME eqis me
Equity Net Issuance scaled by ME eqnetis me
Net Long-Term Debt Issuance
scaled by MEV
Net Short-Term Debt Issuance dstnetis mev
scaled by MEV
Net Debt Issuance scaled by MEV dbnetis mev
Net Issuance scaled by MEV netis mev
Gross Profit scaled by MEV gp mev
Operating Profit after Depreciation ebit mev
scaled by MEV
Sales scaled by MEV sale mev
Operating Cash Flow scaled by ocf mev
MEV
Free Cash Flow scaled by MEV fcf mev
Cash Based Operating Profitability cop mev
scaled by MEV
Financial Cash Flow Change scaled fincf mev
by MEV
Net Income to Sales Quarterly niq saleq std
Volatility
Net Income scaled by Employees ni emp
Sales scaled by Employees sale emp
Net Income scaled by Assets ni at
Quarterly ROE Volatility roeq be std
ROE Volatility roe be std
Gross Product to Assets 5 yr gpoa ch5
Change
ROE 5 yr Change roe ch5
ROA 5 yr Change roa ch5
Operating Cash Flow to Assets 5 yr cfoa ch5
Change

46
 Gross Product to Sales 5 yr Change gmar ch5
Dividend to Price - 1 Month div1m me
Dividend to Price - 3 Months div3m me
Dividend to Price - 6 Months div6m me
Special Dividend to Price - 1 Month divspc1m me
Special Dividend to Price - 12 divspc12m me
Month
Change in Shares - 1 Month chcsho 1m
Change in Shares - 3 Month chcsho 3m
Change in Shares - 6 Month chcsho 6m
Net Equity Payout - 1 Month eqnpo 1m
Net Equity Payout - 3 Month eqnpo 3m
Net Equity Payout - 6 Month eqnpo 6m
Momentum 0-2 Months ret 2 0
Momentum 0-3 Months ret 3 0
Momentum 0-6 Months ret 6 0
Momentum 0-9 Months ret 9 0
Momentum 0-12 Months ret 12 0
Momentum 1-18 Months ret 18 1
Momentum 1-24 Months ret 24 1
Momentum 12-24 Months ret 24 12
Momentum 1-36 Months ret 36 1
Momentum 12-36 Months ret 36 12
Momentum 1-48 Months ret 48 1
Momentum 12-48 Months ret 48 12
Momentum 1-60 Months ret 60 1
Momentum 36-60 Months ret 60 36
Market beta (21 days) beta 21d
Market beta (252 days) beta 252d
Return volatility (252 days) rvol 252d
Idiosyncratic volatility from the ivol capm 60m
CAPM (60 months)
The high-low return volatility rvolhl 21d Corwin and Schultz (2012)

Note: This table shows cluster names as underlined section headings and, for each cluster, a description of the
factors included, the variable name used in the code, the original reference, the sample period used in the original
reference, the sign of the factor (“1” means “long”, “-1” means “short”), and whether the original reference found
the factor to be significant (“1” means “yes”, “0” means “no”). For example, the first value factor “at me” goes
long stocks with high values of assets-to-market and shorts those with low values (and would be done the reverse
if the sign was “-1” instead of “1”).

47
11 Miscellaneous
Table 10: Country Code Key and MSCI Categorization
Country EXCNTRY-Country Code MSCI Categorization
Argentina ARG emerging
Australia AUS developed
Austria AUT developed
Bahrain BHR frontier
Bangladesh BGD frontier
Belgium BEL developed
Bermuda BMU not rated
Botswana BWA standalone
Brazil BRA emerging
Bulgaria BGR standalone
Canada CAN developed
Chile CHL emerging
China CHN emerging
Colombia COL emerging
Croatia HRV frontier
Cyprus CYP not rated
Czech Republic CZE emerging
Denmark DNK developed
Ecuador ECU not rated
Egypt EGY emerging
Estonia EST frontier
Finland FIN developed
France FRA developed
Germany DEU developed
Ghana GHA not rated
Greece GRC emerging
Guernsey GGY not rated
Hong Kong HKG developed
Hungary HUN emerging
Iceland ISL standalone
India IND emerging
Indonesia IDN emerging
Iran, Islamic Republic of IRN not rated
Ireland IRL developed
Israel ISR developed
Italy ITA developed
Jamaica JAM standalone
Japan JPN developed
Jordan JOR frontier
Kazakhstan KAZ frontier
Kenya KEN frontier
Korea, Republic of KOR emerging
Kuwait KWT frontier
Latvia LVA not rated
Lebanon LBN standalone

48
Country EXCNTRY-Country Code MSCI Categorization
Lithuania LTU frontier
Luxembourg LUX not rated
Malawi MWI not rated
Malaysia MYS emerging
Malta MLT standalone
Mauritius MUS frontier
Mexico MEX emerging
Morocco MAR frontier
Namibia NAM not rated
Netherlands NLD developed
New Zealand NZL developed
Nigeria NGA frontier
Norway NOR developed
Oman OMN frontier
Pakistan PAK emerging
Palestinian Territory, Occupied PSE standalone
Peru PER emerging
Philippines PHL emerging
Poland POL emerging
Portugal PRT developed
Qatar QAT emerging
Romania ROU frontier
Russian Federation RUS emerging
Saudi Arabia SAU emerging
Senegal SEN frontier
Serbia SRB frontier
Singapore SGP developed
Slovakia SVK not rated
Slovenia SVN frontier
South Africa ZAF emerging
Spain ESP developed
Sri Lanka LKA frontier
Sweden SWE developed
Switzerland CHE developed
Taiwan, Province of China TWN emerging
Tanzania, United Republic of TZA not rated
Thailand THA emerging
Trinidad and Tobago TTO standalone
Tunisia TUN frontier
Turkey TUR emerging
Uganda UGA not rated
Ukraine UKR standalone
United Arab Emirates ARE emerging
United Kingdom GBR developed
United States USA developed
Uruguay URY not rated
Venezuela, Bolivarian Republic of VEN not rated
Viet Nam VNM frontier
Zambia ZMB not rated

49
 Country EXCNTRY-Country Code MSCI Categorization
Zimbabwe ZWE standalone

References
Abarbanell, J. S., & Bushee, B. J. (1998). Abnormal returns to a fundamental analysis
strategy. Accounting Review , 19–45.
Ali, A., Hwang, L.-S., & Trombley, M. A. (2003). Arbitrage risk and the book-to-market
anomaly. Journal of Financial Economics, 69 (2), 355–373.
Amihud, Y. (2002). Illiquidity and stock returns: cross-section and time-series effects.
Journal of financial markets, 5 (1), 31–56.
Anderson, C. W., & Garcia-Feijoo, L. (2006). Empirical evidence on capital investment,
growth options, and security returns. The Journal of Finance, 61 (1), 171–194.
Ang, A., Chen, J., & Xing, Y. (2006). Downside risk. The review of financial studies, 19 (4),
1191–1239.
Ang, A., Hodrick, R. J., Xing, Y., & Zhang, X. (2006). The cross-section of volatility and
expected returns. The Journal of Finance, 61 (1), 259–299.
Asness, C., Frazzini, A., Gormsen, N. J., & Pedersen, L. H. (2020). Betting against correla-
tion: Testing theories of the low-risk effect. Journal of Financial Economics, 135 (3),
629–652.
Asness, C. S., Frazzini, A., & Pedersen, L. H. (2019). Quality minus junk. Review of
Accounting Studies, 24 (1), 34–112.
Balakrishnan, K., Bartov, E., & Faurel, L. (2010). Post loss/profit announcement drift.
Journal of Accounting and Economics, 50 (1), 20–41.
Bali, T. G., Brown, S. J., & Tang, Y. (2017). Is economic uncertainty priced in the cross-
section of stock returns? Journal of Financial Economics, 126 (3), 471–489.
Bali, T. G., Cakici, N., & Whitelaw, R. F. (2011). Maxing out: Stocks as lotteries and the
cross-section of expected returns. Journal of Financial Economics, 99 (2), 427–446.
Bali, T. G., Engle, R. F., & Murray, S. (2016). Empirical asset pricing: The cross section
of stock returns. John Wiley & Sons.
Ball, R., Gerakos, J., Linnainmaa, J. T., & Nikolaev, V. (2016). Accruals, cash flows,
and operating profitability in the cross section of stock returns. Journal of Financial
Economics, 121 (1), 28–45.
Banz, R. W. (1981). The relationship between return and market value of common stocks.
Journal of financial economics, 9 (1), 3–18.
Barbee Jr, W. C., Mukherji, S., & Raines, G. A. (1996). Do sales–price and debt–equity
explain stock returns better than book–market and firm size? Financial Analysts
Journal , 52 (2), 56–60.
Barth, M. E., Elliott, J. A., & Finn, M. W. (1999). Market rewards associated with patterns
of increasing earnings. Journal of Accounting Research, 37 (2), 387–413.
Basu, S. (1983). The relationship between earnings’ yield, market value and return for nyse
common stocks: Further evidence. Journal of financial economics, 12 (1), 129–156.
Belo, F., & Lin, X. (2012). The inventory growth spread. The Review of Financial Studies,
25 (1), 278–313.

50
Belo, F., Lin, X., & Bazdresch, S. (2014). Labor hiring, investment, and stock return
predictability in the cross section. Journal of Political Economy, 122 (1), 129–177.
Bhandari, L. C. (1988). Debt/equity ratio and expected common stock returns: Empirical
evidence. The journal of finance, 43 (2), 507–528.
Blitz, D., Huij, J., & Martens, M. (2011). Residual momentum. Journal of Empirical
Finance, 18 (3), 506–521.
Bouchaud, J.-P., Krueger, P., Landier, A., & Thesmar, D. (2019). Sticky expectations and
the profitability anomaly. The Journal of Finance, 74 (2), 639–674.
Boudoukh, J., Michaely, R., Richardson, M., & Roberts, M. R. (2007). On the importance
of measuring payout yield: Implications for empirical asset pricing. The Journal of
Finance, 62 (2), 877–915.
Bradshaw, M. T., Richardson, S. A., & Sloan, R. G. (2006). The relation between corporate
financing activities, analysts’ forecasts and stock returns. Journal of accounting and
economics, 42 (1-2), 53–85.
Brennan, M. J., Chordia, T., & Subrahmanyam, A. (1998). Alternative factor specifications,
security characteristics, and the cross-section of expected stock returns. Journal of
financial Economics, 49 (3), 345–373.
Chan, L. K., Lakonishok, J., & Sougiannis, T. (2001). The stock market valuation of research
and development expenditures. The Journal of finance, 56 (6), 2431–2456.
Chordia, T., Subrahmanyam, A., & Anshuman, V. R. (2001). Trading activity and expected
stock returns. Journal of financial Economics, 59 (1), 3–32.
Cooper, M. J., Gulen, H., & Schill, M. J. (2008). Asset growth and the cross-section of stock
returns. the Journal of Finance, 63 (4), 1609–1651.
Corwin, S. A., & Schultz, P. (2012). A simple way to estimate bid-ask spreads from daily
high and low prices. The Journal of Finance, 67 (2), 719–760.
Daniel, K., & Titman, S. (2006). Market reactions to tangible and intangible information.
The Journal of Finance, 61 (4), 1605–1643.
Datar, V. T., Naik, N. Y., & Radcliffe, R. (1998). Liquidity and stock returns: An alternative
test. Journal of financial markets, 1 (2), 203–219.
De Bondt, W. F., & Thaler, R. (1985). Does the stock market overreact? The Journal of
finance, 40 (3), 793–805.
Dechow, P. M., Sloan, R. G., & Soliman, M. T. (2004). Implied equity duration: A new
measure of equity risk. Review of Accounting Studies, 9 (2), 197–228.
Desai, H., Rajgopal, S., & Venkatachalam, M. (2004). Value-glamour and accruals mispric-
ing: One anomaly or two? The Accounting Review , 79 (2), 355–385.
Dichev, I. D. (1998). Is the risk of bankruptcy a systematic risk? the Journal of Finance,
53 (3), 1131–1147.
Dimson, E. (1979). Risk measurement when shares are subject to infrequent trading. Journal
of Financial Economics, 7 (2), 197–226.
Fairfield, P. M., Whisenant, J. S., & Yohn, T. L. (2003). Accrued earnings and growth:
Implications for future profitability and market mispricing. The accounting review ,
78 (1), 353–371.
Fama, E. F., & French, K. R. (1992). The cross-section of expected stock returns. the
Journal of Finance, 47 (2), 427–465.

51
Fama, E. F., & French, K. R. (1997). Industry costs of equity. Journal of Financial Eco-
nomics, 43 (2), 153-193. Retrieved from https://www.sciencedirect.com/science/
article/pii/S0304405X96008963 doi: https://doi.org/10.1016/S0304-405X(96)
00896-3
Fama, E. F., & French, K. R. (2015). A five-factor asset pricing model. Journal of financial
economics, 116 (1), 1–22.
Fama, E. F., & MacBeth, J. D. (1973). Risk, return, and equilibrium: Empirical tests.
Journal of political economy, 81 (3), 607–636.
Foster, G., Olsen, C., & Shevlin, T. (1984). Earnings releases, anomalies, and the behavior
of security returns. Accounting Review , 574–603.
Francis, J., LaFond, R., Olsson, P. M., & Schipper, K. (2004). Costs of equity and earnings
attributes. The accounting review , 79 (4), 967–1010.
Frankel, R., & Lee, C. M. (1998). Accounting valuation, market expectation, and cross-
sectional stock returns. Journal of Accounting and economics, 25 (3), 283–319.
Frazzini, A., & Pedersen, L. H. (2014). Betting against beta. Journal of Financial Economics,
111 (1), 1–25.
George, T. J., & Hwang, C.-Y. (2004). The 52-week high and momentum investing. The
Journal of Finance, 59 (5), 2145–2176.
Hafzalla, N., Lundholm, R., & Matthew Van Winkle, E. (2011). Percent accruals. The
Accounting Review , 86 (1), 209–236.
Hahn, J., & Lee, H. (2009). Financial constraints, debt capacity, and the cross-section of
stock returns. The Journal of Finance, 64 (2), 891–921.
Harvey, C. R., & Siddique, A. (2000). Conditional skewness in asset pricing tests. The
Journal of finance, 55 (3), 1263–1295.
Haugen, R. A., & Baker, N. L. (1996). Commonality in the determinants of expected stock
returns. Journal of financial economics, 41 (3), 401–439.
Heston, S. L., & Sadka, R. (2008). Seasonality in the cross-section of stock returns. Journal
of Financial Economics, 87 (2), 418–445.
Hirshleifer, D., Hou, K., Teoh, S. H., & Zhang, Y. (2004). Do investors overvalue firms with
bloated balance sheets? Journal of Accounting and Economics, 38 , 297–331.
Hou, K., Xue, C., & Zhang, L. (2015). Digesting anomalies: An investment approach. The
Review of Financial Studies, 28 (3), 650–705.
Huang, A. G. (2009). The cross section of cashflow volatility and expected stock returns.
Journal of Empirical Finance, 16 (3), 409–429.
Jegadeesh, N. (1990). Evidence of predictable behavior of security returns. The Journal of
finance, 45 (3), 881–898.
Jegadeesh, N., & Livnat, J. (2006). Revenue surprises and stock returns. Journal of
Accounting and Economics, 41 (1-2), 147–171.
Jegadeesh, N., & Titman, S. (1993). Returns to buying winners and selling losers: Implica-
tions for stock market efficiency. The Journal of finance, 48 (1), 65–91.
Jensen, T. I., Kelly, B. T., & Pedersen, L. H. (2022). Is there a replication crisis in finance?
Journal of Finance, forthcoming.
Jiang, G., Lee, C. M., & Zhang, Y. (2005). Information uncertainty and expected returns.
Review of Accounting Studies, 10 (2-3), 185–221.

52
Lakonishok, J., Shleifer, A., & Vishny, R. W. (1994). Contrarian investment, extrapolation,
and risk. The journal of finance, 49 (5), 1541–1578.
Lamont, O., Polk, C., & Saaá-Requejo, J. (2001). Financial constraints and stock returns.
The review of financial studies, 14 (2), 529–554.
Lev, B., & Nissim, D. (2004). Taxable income, future earnings, and equity values. The
accounting review , 79 (4), 1039–1074.
Li, F. (2011). Earnings quality based on corporate investment decisions. Journal of Ac-
counting Research, 49 (3), 721–752.
Litzenberger, R. H., & Ramaswamy, K. (1979). The effect of personal taxes and dividends
on capital asset prices: Theory and empirical evidence. Journal of financial economics,
7 (2), 163–195.
Liu, W. (2006). A liquidity-augmented capital asset pricing model. Journal of financial
Economics, 82 (3), 631–671.
Loughran, T., & Wellman, J. W. (2011). New evidence on the relation between the enterprise
multiple and average stock returns. Journal of Financial and Quantitative Analysis,
1629–1650.
Lyandres, E., Sun, L., & Zhang, L. (2008). The new issues puzzle: Testing the investment-
based explanation. The Review of Financial Studies, 21 (6), 2825–2855.
Miller, M. H., & Scholes, M. S. (1982). Dividends and taxes: Some empirical evidence.
Journal of Political Economy, 90 (6), 1118–1141.
Novy-Marx, R. (2011). Operating leverage. Review of Finance, 15 (1), 103–134.
Novy-Marx, R. (2012). Is momentum really momentum? Journal of Financial Economics,
103 (3), 429–453.
Novy-Marx, R. (2013). The other side of value: The gross profitability premium. Journal
of financial economics, 108 (1), 1–28.
Ortiz-Molina, H., & Phillips, G. M. (2014). Real asset illiquidity and the cost of capital.
Journal of Financial and Quantitative Analysis, 1–32.
Palazzo, B. (2012). Cash holdings, risk, and expected returns. Journal of Financial Eco-
nomics, 104 (1), 162–185.
Penman, S. H., Richardson, S. A., & Tuna, I. (2007). The book-to-price effect in stock
returns: accounting for leverage. Journal of accounting research, 45 (2), 427–467.
Piotroski, J. D. (2000). Value investing: The use of historical financial statement information
to separate winners from losers. Journal of Accounting Research, 1–41.
Pontiff, J., & Woodgate, A. (2008). Share issuance and cross-sectional returns. The Journal
of Finance, 63 (2), 921–945.
Richardson, S. A., Sloan, R. G., Soliman, M. T., & Tuna, I. (2005). Accrual reliability,
earnings persistence and stock prices. Journal of accounting and economics, 39 (3),
437–485.
Rosenberg, B., Reid, K., & Lanstein, R. (1985). Persuasive evidence of market inefficiency.
The Journal of Portfolio Management, 11 (3), 9–16.
Sloan, R. G. (1996). Do stock prices fully reflect information in accruals and cash flows
about future earnings? Accounting review , 289–315.
Soliman, M. T. (2008). The use of dupont analysis by market participants. The Accounting
Review , 83 (3), 823–853.

53
Stambaugh, R. F., & Yuan, Y. (2017). Mispricing factors. The Review of Financial Studies,
30 (4), 1270–1315.
Thomas, J., & Zhang, F. X. (2011). Tax expense momentum. Journal of Accounting
Research, 49 (3), 791–821.
Thomas, J. K., & Zhang, H. (2002). Inventory changes and future returns. Review of
Accounting Studies, 7 (2), 163–187.
Titman, S., Wei, K. C. J., & Xie, F. (2004, December). Capital investments and stock
returns. The Journal of Financial and Quantitative Analysis, 39 (4), 677-700.
Xie, H. (2001). The mispricing of abnormal accruals. The accounting review , 76 (3), 357–373.

54