Unlimited loop of events tested when using eventstudy2

[Daniel Handojo]

Hello!

I am currently conducting my final thesis and for that I am using the package eventstudy2, as this has the most comprehensive statistics.

However, I am not conducting an Abnormal Return event study, but an Abnormal Trading Volume event study. In Abnormal Trading Volume the ln(Volume Turnover) is taken instead of the Returns. However, the calculation is exactly the same using the Market Model.


When using the package, I transformed my input .dta exactly as the example given in the help page. But, it seems that my event study is stuck in an infinite loop of events, even though I only have 3 events in order to test the syntax.

I feel like I tried already everything and I cannot see where the error is.

My Code:

use 01_request_mutated.dta, clear        

eventstudy2 Security_id Date using 02_firm_mutated, ret(Return) car1LB(-1) car1UB(1) mod(FM) marketfile(03_market_mutated) mar(MKT) idmar(Market_reference) aar(sol_output_aar) car(sol_output_car) ar(sol_output_ar) crossfile(sol_output_crossfile) diag(sol_output_diag) graph(sol_output_graph)

I have 3 files:
01_request_mutated.dta are the event dates and is stored in memory (Contains: Date, Market_reference, Security_id)
02_firm_mutated.dta is a list of the trade turnover for my firms (Contains: Date, Market_reference, Security_id, Return)
03_market_mutated.dta is the list of the benchmark market index (Contains: Date, Market_reference, MKT)


Many thanks for every help!

Best regards,
Daniel Handojo

[Willemijn Olree]

Hi Daniel,

It is a long time ago, but I'm having the same problem right now. Could you maybe tell me if, and how you solved this problem? Or is there someone else who can help me? Many thanks!

Greetz

[Thomas Kaspereit]

Dear Sir,

Could you please use returns in decimal format, e.g. a daily ten percent return is 0.1

Best Thomas

[Tian Luo]

Dear all,

It seems a long time, but I am having the same problem. I changed my returns in decimal format but only some data sets work. For some others, it just keeps in infinite looping. Is there anybody know the reason? Many thanks!

Best,
Tian

[Jing Xu Xu]

I ran the event study using eventstudy2 on one index and it worked fine. Then I used an alternative index, which has 9 more missing value out of 379 observations than the first index. I came across same issue described in this post. I "fixed" the issue by allowing more observations in my data.

My returns are always in decimal format. 10% return is 0.1.