Timeline for Generalized difference-in-difference: pre-period & pre-trend
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 24, 2021 at 5:41 | comment | added | Thomas Bilach | Event study estimation is easier in this setting. One way to proceed is to interact the ESG dummies with a series of time indicators for the periods before and after the crisis. Authors will typically plot the coefficients on some of these interactions to see how they evolve over time. | |
| Nov 23, 2021 at 9:09 | vote | accept | Rudolf | ||
| Nov 19, 2021 at 13:28 | comment | added | Rudolf | Thank you very much for the clear and detailed explanation. Aside from the visual representation as you recommend for testing the parallel trend, is there another formal test for validating the parallel trend? For example, I have seen authors use event studies. Are these tests suitable for this setting, and which test would you recommend? | |
| Nov 19, 2021 at 6:12 | history | edited | Thomas Bilach | CC BY-SA 4.0 |
Minor edits.
|
| Nov 19, 2021 at 3:10 | history | edited | Thomas Bilach | CC BY-SA 4.0 |
Minor textual edits.
|
| Nov 19, 2021 at 2:40 | history | answered | Thomas Bilach | CC BY-SA 4.0 |