Fama-French Three-Factor

Analysis
Learning Objectives

Upon completing this assignment, students will be able

to:

• Describe the Fama-French three-factor model

• Build a regression model in Excel

• Analyze how well an asset’s returns are explained by

the Fama-French three-factor model
 From a Single to Multi-Factor Model
Ri = asset return
• The traditional regression
Rf = risk-free rate of return
model for analyzing excess
returns is the Capital Asset
𝛼i = difference between the asset’s
Pricing Model (CAPM), a return and the expected return
single-factor model:
βi = measure of the asset’s
(𝑅𝑖−𝑅𝑓) = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑀𝐾𝑇−𝑅𝑓) + ei systematic risk in relation to the
market
• The model stipulates
there is only one risk RMKT = the market return
factor: the return on
the market portfolio. ei = random error/non-systematic risk
From a Single to Multi-Factor Model
(cont.)
• According to the CAPM theory, in the previous slide’s
equation, 𝛼 (alpha) should equal 0. However, that is not
always the case.

• Researchers began investigating models where additional

factors (e.g. interest rates) could be added to the market
factor in order to better explain excess returns.

• Analyzing historical data, Eugene Fama and Kenneth French

noticed that small-cap stocks and value stocks tended to
outperform large-cap stocks and growth stocks.
Fama-French Three-Factor Model
• Fama and French added a size factor and a value factor to
the market factor:

(𝑅𝑖−𝑅𝑓) = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑀𝐾𝑇−𝑅𝑓)+ si(SMB) + hi(HML) + ei

• Empirically, their three-factor model explained historical

returns better than the single-factor market model,
explaining over 90% of excess returns as opposed to
approximately 70%.
Size: (SMB) Small Minus Big
(𝑅𝑖−𝑅𝑓) = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑀𝐾𝑇−𝑅𝑓)+ si(SMB) + hi(HML) + ei

• The size premium (SMB) is the average monthly return

on the smallest 30% of stocks (in terms of market
capitalization) minus the average monthly return on the
largest 30%.

• When small stocks do well relative to large stocks, this

will be positive; when they do worse than large stocks,
this will be negative.
Value: (HML) High Minus Low
(𝑅𝑖−𝑅𝑓) = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑀𝐾𝑇−𝑅𝑓)+ si(SMB) + hi(HML) + ei

• The value premium (HML) is the average monthly return

for the 50% of stocks with the highest book-to-market ratio
minus the average return for the 50% of stocks with the
lowest book-to-market ratio.

• When high value stocks do well relative to low value stocks,

this will be positive; when they do worse than low value
stocks, this will be negative.

• High book-to-market stocks are considered “value” stocks;

low book-to-market stocks are considered “growth” stocks.
Factor Betas
• Extending the single factor CAPM model, the Fama-French
model uses three factor betas:
(𝑅𝑖−𝑅𝑓) = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑀𝐾𝑇−𝑅𝑓)+ si(SMB) + hi(HML) + ei

• A factor beta (sometimes called a “factor loading”) is the

sensitivity of security’s returns to a particular systematic
factor.

• With this model, market returns can roughly be

explained by three factors: (1) exposure to the overall
market (RMKT - Rf); (2) exposure to small cap stocks
(SMB); and (3) exposure to value stocks (HML).
The Fama-French Factors
• Ken French publishes datasets of the Fama-French factors
for distribution from his web site at Dartmouth University.

• For more detailed information on how these factors were

calculated, visit his web site at:

http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/d
ata_library.html

• As research continues, additional factors continue to be

added to the original three-factor model.
 Assignment
• Use a WRDS query to select and download historical
monthly returns for two exchange-traded funds (ETFs).

• The Fama-French three-factor data will be provided in this

query’s output.

• After downloading the data, use the regression function in

Excel to perform a multiple linear regression for each fund’s
excess returns using the Fama-French three-factor data as
the three independent variables.

• Compare the results for the two funds.

About Excel
• The following instructions were created using Excel
2013 (Microsoft Office Professional Plus 2013) using a
PC.

• Depending on your version of Excel, screens may look

slightly different, and the corresponding steps may
need to be adjusted.

• If you are using a different version of Excel and need

additional help, access the Microsoft Excel Help Center
at: https://support.office.com/en-us/excel
 Checking Data Analysis in Excel
• Begin by checking to
make sure your Excel
application has the Data
Analysis functionality
loaded.

• To check, select the

Data tab on the ribbon.

• You should see the Data

Analysis button in the
Analysis group at the
top right corner of the
screen.
 Installing Data Analysis in Excel
If your Excel does not have Data Analysis, use these
installation steps from Microsoft Office support:

1. Click the File tab.

2. Click Options.

3. Click Add-Ins.

4. In the Add-Ins box, select the Analysis ToolPak, and then

click OK.
ETF Returns in WRDS

• Exchange-traded funds (ETFs) are traded on stock

exchanges. To download ETF returns, use the WRDs
Stock Return query at the following link:

https://wrds-
www.wharton.upenn.edu/classroom/stock-and-factor-
returns/
 ETF Returns Query
• Enter the following ETF
tickers:

VTV
VBK

• Enter the date range for

this assignment: 2010 –
2015.

• Click Submit to run the

query.
 ETF Returns Query (cont.)
• Scroll to the bottom of
the screen and click
Download.

• After opening the file in

Excel, you may need to
click Enable Editing to
allow you to continue
the assignment.
 Removing Extra Data
• The WRDS query was designed to provide data to build
several types of models.

• For this Fama-French three-factor analysis, only some

of this data is relevant.
 Removing Extra Data (cont.)
• Delete the following three columns of data that will not
be used in this model: MKT COMPOSITE RETURN, S&P
RETURN, and MOMENTUM FACTOR.
Arranging the Worksheet
• Move the RISK-FREE RATE column to a new location
between the INDEX FUNDS: VANGUARD SMALL-CAP GRWTH
ETF and the VANGUARD VALUE ETF columns.

• This new arrangement, which keeps the three Fama-French

factors in a row, will help simplify the regression process.
Subtract the Risk-Free Rate
• Add a new column after the Risk-Free Rate and label it
Vanguard Small-Cap Growth ETF – Rf.

• In this column’s first cell, create a formula to subtract the

risk-free rate from Vanguard’s Small-Cap Growth ETF
returns. The syntax for the formula can be found in the
screenshot below.
Subtract the Risk-Free Rate (cont.)
• Click Enter. The value of the fund’s returns, less the risk-
free rate, appears.

• Populate the rest of the column with the formula by

double-clicking the cell’s bottom right corner.
Access the Regression Function
• Select the Data tab.

• Click Data Analysis.

 Access the Regression Function
(cont.)
• Select Regression in
the Analysis Tools
window and click OK.
 The Dependent Variable
• The Regression Input
window appears. Select the
Labels checkbox to retain
the column’s labels in your
results.

• The Y-axis is the dependent

variable. This is the ETF’s
excess returns:

(Fund Return - Risk Free

Rate)
The Dependent Variable (cont.)
• In the Regression window,
place your cursor in the
Input Y Range field.

• In the spreadsheet, click first

cell of the Vanguard Small-
Cap Growth Fund – Rf
column.

• Drag downward until you

reach the last month of data.
This is the Input Y Range.
 The Independent Variables
• In the Regression window,
place your cursor in the
Input X Range field.

• The X-axis is the independent variable. Remember, in the Fama-

French 3-Factor model, there are three independent variables:

1. Market: (RMKT - Rf) = (Market Return - Risk Free Rate)

2. Size: SMB = Small Minus Big

3. Value: HML = High Minus Low

 The Independent Variables (cont.)
• In the spreadsheet, use
your cursor to select all
three labeled cells of the
Fama-French Factors.

• Drag downward until you

reach the last month of
data. These three columns
represent the Input X
Range.
Calculate the Regression
• Check the Input
Ranges to make sure
you captured the full
range of data, from
rows 1 to 73.

• Click OK to calculate
the regression.
Optional: Turn off Scientific
Notation in Excel Results
• A new worksheet
appears, with the
regression results.

• Note the letter “E” in

some of the value
fields. For values with
more than 15 digits,
Excel displays them,
by default, using
scientific notation.
Optional: Turn off Scientific
Notation in Excel Results (cont.)
• To simplify your
results for analysis,
you may wish to turn
off scientific notation.
Use your cursor to
select all the columns
with numerical data.

• Right click to open the

options and select
Format Cells.
 Optional: Turn off Scientific
Notation in Excel Results (cont.)
• Select Number in the
Number tab.

• Leave the default

number of decimal
places at 2 and click OK.

• Your results will now be

rounded to the nearest
2 decimal places. Keep
in mind some precision
is now lost.
 Summary Output: Interpreting R2

• R Square provides information on the

explanatory power of the linear
regression model; it indicates how well
the data “fits” the model.

• The Adjusted R Square is modified to

adjust for the number of independent
variables in the model, and is therefore
considered the more conservative
estimate.
 Interpreting 2
R (cont.)

• Here, the Adjusted R Square measures

the degree to which this ETF’s excess
returns can be attributed to the
independent variables.

• During this time period, 97% of this

ETF’s excess returns can be attributed
to the three Fama-French factors.
 Interpreting the Regression Data
• The intercept is the alpha, and the three subsequent
coefficients are the beta factor values.

• The results
can be
rewritten as:
 Interpreting the Regression Data
(cont.)
• As you interpret the fund’s sensitivity to the three factors,
keep in mind that this Vanguard ETF is designed to “tilt”
toward small cap growth stocks.
 Comparing ETF Regression Data

• Repeat the process to perform a regression analysis on the

second ETF, ticker VTV.

• Remember to subtract the risk-free rate to get the excess

returns.

• As you interpret this fund’s sensitivity to the factors, keep in

mind that Vanguard has designed it to skew toward large-cap
value stocks.
Compare the ETFs in Terms of the
Fama-French Factors

Vanguard Small-Cap Growth ETF

Vanguard Large-Cap Value ETF

Conclusion
• In the Capital Asset Pricing Model, the market portfolio
return is the sole source of risk.

• The Fama-French three-factor model suggests that portfolio

returns can roughly be explained by three factors:

(1) exposure to the broad market (RMKT - Rf)

(2) exposure to small stocks (SMB)
(3) exposure to value stocks (HML)

• A factor beta represents the sensitivity of a portfolio’s

returns to changes in a systematic factor.
Notes (cont.)
For Fama-French data:

• Monthly returns on the market portfolio are value-

weighted returns for all firms listed on the NYSE, AMEX,
and NASDAQ.

• The risk-free rate is the return on 1-month T-bills.

• Additional information is available on Ken French’s web

site:
http://mba.tuck.dartmouth.edu/pages/faculty/ken.french
/index.html