Questions on the Eventstudy 2 command

[Philip HU]

Dear Stata Users,

I am using the Evenstudy2 command to perform an event study for a portfolio of stock returns to a common policy event.

I am using code like this:

use Event_Stud.dta, clear

eventstudy2 ISIN day_year using Return_File, returns(exret) model(FM) marketfile(French_5_Factors) marketreturn(mktrf) factor1(smb) factor2(hml) factor3(rmw) factor4(cma) riskfreerate(rf) evwlb(-10) evwub(10) car1LB(-10) car1UB(-5) car2LB(-5) car2UB(0) car3LB(0) car3UB(5)  car4LB(5) car4UB(10)  eswlb(-250) eswub(-30) replace

I have two question on the command:

• Why do the -10 to -5 CAAR do not equal the AARs that I average by hand from t=-10 to t=-5 ?
• Is there a way to add confidence intervals to both the AARs and CAAR estimates?

Thank you so much for helping!

[Thomas Kaspereit]

Dear Sir,

Regarding the question:

“Why do the −10 to −5 CAAR not equal the AARs that I average by hand from t = −10 to t = −5?”

please refer to the following discussion:
https://www.statalist.org/forums/for...ng-eventstudy2

Regarding the question:

“Is there a way to add confidence intervals to both the AARs and CAAR estimates?”

The output files already provide all relevant test statistics. Together with the appropriate distributional assumptions and the corresponding degrees of freedom, this information should allow you to construct the confidence intervals.

Best regards,
Thomas Kaspereit