DID in Stata: Difference-in-Differences Stata Tutorial

1. Using the "basic" method

2. Using the "hashtag" method

3. Using the "diff" command

4. Useful Resources

• Getting sample data.

use "https://dss.princeton.edu/training/Panel101.dta", clear

• Create a dummy variable to indicate the time when the treatment started. Let's assume that the treatment started in 1994. In this case, years before 1994 will have a value of 0, and years from 1994 onward a 1.

gen time = (year>=1994) & !missing(year)

• Create a dummy variable to identify the group exposed to the treatment. In this example, let's assume that countries with codes 5, 6, and 7 were treated (=1). Countries 1-4 were not treated (=0).

gen treated = (country>4) & !missing(country)

• Create an interaction between time and treated. We will call this interaction ‘did’

gen did = time*treated

• 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 |  Coefficient    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 difference-in-differences estimator. The effect is significant at 10% level, with the treatment having a negative effect.

• No need to generate interaction while using the hashtag method. Estimate using the following command

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  |  Coefficient    std. err.      t      p>|t|      [95% conf. interval]

-----------------------------------------------------------------------------------

     1.time |  2.29e+09       9.00e+08      2.54     0.013      4.92e+08   4.09e+09

  1.treated |  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 difference-in-differences estimator (‘did’ in the previous example). The effect is significant at 10%, with the treatment having a negative effect.

• The command diff is user‐defined for Stata. To install, type

ssc install diff

• Estimating using the diff command

diff y, t(treated) p(time)

Note: "treated" and "time" in parentheses are dummies for treatment and time; see the "basic" method

. 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

---------------------------------------------------------------

 Outcome var.   |  y       | S. Err.   |  t       | P>|t|

---------------------------------------------------------------

Baseline

Control    |  3.6e+08 |           |         |

Treated    |  2.1e+09 |           |         |

Diff (T-C) |  1.8e+09 |  1.1e+09  | 1.58    | 0.120

Follow-up

Control    |  2.6e+09 |           |         |

Treated    |  1.9e+09 |           |         |

Diff (T-C) | -7.4e+08 |  9.2e+08  | -0.81   | 0.422

Diff-in-Diff    | -2.5e+09 |  1.5e+09  | -1.73   | 0.088* 

------------------------------------------------------------------

R-square:    0.08
* Means and Standard Errors are estimated by linear regression
**Inference: *** p<0.01; ** p<0.05; * p<0.1

• The coefficient for ‘Diff-in-Diff’ is the difference-in-differences estimator. The effect is significant at 10% level, with the treatment having a negative effect.

** Type help diff for more details/options