DifferencesinDifferences
(usingStata)
(workinprogress)
OscarTorresReyna
otorres@[Link]
August 2015 [Link]
Differenceindifferences(DID)
Estimationstepbystep
* Getting sample data.
use "[Link] clear
* Create a dummy variable to indicate the time when the treatment started. Lets
assume that treatment started in 1994. In this case, years before 1994 will
have a value of 0 and 1994+ a 1. If you already have this skip this step.
gen time = (year>=1994) & !missing(year)
* Create a dummy variable to identify the group exposed to the treatment. In
this example lets assumed that countries with code 5,6, and 7 were treated
(=1). Countries 1-4 were not treated (=0). If you already have this skip this
step.
gen treated = (country>4) & !missing(country)
* Create an interaction between time and treated. We will call this interaction
did
gen did = time*treated
OTR 2
Differenceindifferences(DID)
Estimationstepbystep
* Estimating the DID estimator
reg y time treated did, r
. reg y time treated did, r
Linear regression Number of obs = 70
F(3, 66) = 2.17
Prob > F = 0.0998
R-squared = 0.0827
Root MSE = 3.0e+09
Robust
y Coef. Std. Err. t P>|t| [95% Conf. Interval]
time 2.29e+09 9.00e+08 2.54 0.013 4.92e+08 4.09e+09
treated 1.78e+09 1.05e+09 1.70 0.094 -3.11e+08 3.86e+09
did -2.52e+09 1.45e+09 -1.73 0.088 -5.42e+09 3.81e+08
_cons 3.58e+08 7.61e+08 0.47 0.640 -1.16e+09 1.88e+09
* The coefficient for did is the differences-in-differences
estimator. The effect is significant at 10% with the treatment having
a negative effect.
OTR 3
Differenceindifferences(DID)
Estimationstepbystep
* Estimating the DID estimator (using the hashtag method, no need to
generate the interaction)
reg y time##treated, r
. reg y time##treated, r
Linear regression Number of obs = 70
F(3, 66) = 2.17
Prob > F = 0.0998
R-squared = 0.0827
Root MSE = 3.0e+09
Robust
y Coef. Std. Err. t P>|t| [95% Conf. Interval]
[Link] 2.29e+09 9.00e+08 2.54 0.013 4.92e+08 4.09e+09
[Link] 1.78e+09 1.05e+09 1.70 0.094 -3.11e+08 3.86e+09
time#treated
1 1 -2.52e+09 1.45e+09 -1.73 0.088 -5.42e+09 3.81e+08
_cons 3.58e+08 7.61e+08 0.47 0.640 -1.16e+09 1.88e+09
* The coefficient for time#treated is the differences-in-
differences estimator (did in the previous example). The effect is
significant at 10% with the treatment having a negative effect.
OTR 4
Differenceindifferences(DID)
Usingthecommanddiff
[Link]
ssc install diff Dummiesfortreatmentandtime,seepreviousslide
. diff y, t(treated) p(time)
Number of observations in the DIFF-IN-DIFF: 70
Baseline Follow-up
Control: 16 24 40
Treated: 12 18 30
28 42
R-square: 0.08273
DIFFERENCE IN DIFFERENCES ESTIMATION
BASE LINE FOLLOW UP
Outcome Variable Control Treated Diff(BL) Control Treated Diff(FU) DIFF-IN-DIFF
y 3.6e+08 2.1e+09 1.8e+09 2.6e+09 1.9e+09 -7.4e+08 -2.5e+09
Std. Error 7.4e+08 8.5e+08 1.1e+09 6.0e+08 7.0e+08 9.2e+08 1.5e+09
t 0.49 3.6e+08 1.58 3.6e+08 4.4e+09 1.8e+09 -1.73
P>|t| 0.629 0.015 0.120 0.000 0.008 0.422 0.088*
* Means and Standard Errors are estimated by linear regression
**Inference: *** p<0.01; ** p<0.05; * p<0.1
pvalueforthetreatment
Typehelp diff formoredetails/options OTR effect,orDIDestimator.5
References
Introductiontoeconometrics,[Link],[Link].2nd
ed.,Boston:PearsonAddisonWesley,2007.
DifferenceinDifferencesEstimation,Imbens/Wooldridge,Lecture
Notes10,summer2007.
[Link]
Lecture3:DifferencesinDifferences,FabianWaldinger
[Link]
ching/ec9a8/slides/lecture_3__did.pdf
OTR 6
