st: Re: statalist-digest V4 #3134

[Kit Baum <baum@bc.edu>]

<  >

You should have a look at

http://dss.princeton.edu/usingdata/stata/analysis/eventstudy.html

and per the Statalist FAQ, please do not send WinJunk to the  
Statalist (winmail.dat, etc.) Configure your email program to use  
plain ASCII text mail without vcards, winmail, etc. If your corporate  
environment does not permit this, use a free email account like gmail  
to send to Statalist (but subscribe it first).


Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Jul 27, 2008, at 02:33 , Kanika wrote:


> I am planning to do an event study for the first time and need some  
> advic=
> e.
>
> Before I ask my question, here is what I plan to test-- How does  
> stock ma=
> rket reacts to rating announcements?
>
> For my study I have a small sample of 29 companies and there are  
> two anno=
> uncements made till now. I have data on these companies from the  
> year 200=
> 5 till date. I have defined the estimation and event window (3 day  
> event =
> window) and calculated abnormal returns for the event windows. Then  
> I cal=
> culated cumulative abnormal return by aggregating abnormal return  
> over th=
> e window. So basically, I have two cumulative abormal return for  
> each com=
> pany (for two announcement). Does the methodology sounds fine?  
> Please adv=
> ice.
>
> I am not sure how to test the significance of these abnormal  
> returns?=20
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/