search

chaisemartinPackages / did_multiplegt_dyn

|| Stata | R || Estimation of event-study Difference-in-Difference (DID) estimators in designs with multiple groups and periods, and with a potentially non-binary treatment that may increase or decrease multiple times.
38 stars 8 forks source link

Test of joint nullity of the effects and Test of equality of the effects: what is the difference? #83

Closed LuigiBiagini closed 1 month ago

LuigiBiagini commented 2 months ago

Sorry for this stupid question but I don't understand the difference between the "Test of joint nullity of the effects" and "Test of equality of the effects"?

chaisemartinPackages commented 1 month ago

Dear Luigi,

The first one (T1) tests whether all estimated effects are jointly 0. H_01 is : ß_l == 0 for all l >0 The second (T2) tests whether all estimated effects are equal. H_02 is : ß_l == ß_k for all 0 < l,k <= max effect

They test for 2 different situations. T1 allows to identify if you treatment has any detectable (statistically significant) effect. T2 allows to detect a static treatment effect (whether the effect is not dynamic). What might be misguiding is that H_01 (all effects are nil) implies H_02. However, when you have no effect at all, it seems a bit silly to test whether those effects are time-varying..

Hope this helps, Romain

lbiagini75 commented 1 month ago

Thank you for your kind and very precise response. I have read the dCDH2024 paper but I need to have other clarifications to allow aligning the R output with the paper. This response allows me to overcome my doubts.