1

I have the following DiD model: $y = \beta_0 + \beta_1Treat + \beta_2Post + \beta_3(Treat*Post) + \epsilon$

Where the outcome variable $y$ is expressed as a daily percent change (can be negative or positive). Should the outcome variable be transformed as a ratio as explained here?

What would be the correct interpretation of the $\beta_3$ coefficient if it is negative?

Sam Kuhn
  • 11
  • It would simply mean that the intervention has a negative effect on y and is leading to a negative daily percent change in your variable. More information on what your variables actually are would be helpful. Cant really say much about whether you can use ratios or some other transformation if we don't know what the variables are. – Spur Economics Jul 20 '22 at 06:08

0 Answers0