Event Study and CAR

[Toby Jones]

Hi all,

After looking at previous ES/CAR threads and a popular Princeton link (https://dss.princeton.edu/online_hel...study.html#car) I am still struggling to create my event study. What I am looking to achieve is to assess if there are CAR regarding events where the final column "ecbannouncement" equals 1. The data are for two indexes, or equivalently 2 firms if you prefer. There are multiple events. I have price data, and % change from one price to the next. I only want to measure the change on the day of the event. One of the key issues I have is that my events are randomly scattered, and so I cannot use an estimation window 'rule'.

I am working in Stata 13.0. Observations 5000.

Any help would be much appreciated.

input str10 date float(iboxx iboxx_1d_percent_change eurostoxx eurostx_1d_percent_change) byte ecbannouncement
"05/06/2018"   390.5586    .3450298 386.89  -.31 0
"04/06/2018"  389.2157  -.32660925 388.11   .31 0
"01/06/2018"  390.4911     .244412 386.91  1.01 0
"31/05/2018"   389.539   .16564375 383.06  -.63 0
"30/05/2018"  388.8948  -.50721276 385.49   .27 0
"29/05/2018"  390.8774    .4338071 384.47 -1.37 0
"25/05/2018"  389.1891    .3527581 391.08   .14 0
"24/05/2018"   387.821    .2287407 390.54  -.52 0
"23/05/2018"  386.9359    .3488361 392.58  -1.1 0
"22/05/2018" 385.59085   -.2346069 396.94   .27 0
"21/05/2018"  386.4976   -.4132477 395.87    .3 0
"18/05/2018"  388.1014   .21049033 394.67  -.28 0
"17/05/2018"  387.2862   -.1688323 395.79   .66 0
"16/05/2018"  387.9412    .3571809 393.21   .21 0
"15/05/2018"  386.5605   .24801083 392.37   .05 0
"14/05/2018"  385.6041  -.07222659 392.19  -.05 0
"11/05/2018"  385.8828  -.10083818  392.4   .11 0
"10/05/2018"  386.2723   -.6410539 391.97  -.12 0
"09/05/2018"  388.7645   .07585402 392.44   .63 0
"08/05/2018" 388.46985    .4058572    390   .13 0
"04/05/2018"  386.8996  -.14886068 387.03   .63 0
"03/05/2018"  387.4764   .07181849 384.62  -.73 0
"02/05/2018"  387.1983  -.14353397 387.44   .63 0
"01/05/2018"  387.7549  -.24312073 385.03  -.08 0
"30/04/2018"  388.6999    .2201802 385.32   .18 0
"27/04/2018"  387.8459   -.6185774 384.64   .23 0
"26/04/2018"    390.26     .664869 383.75   .94 1

[Phil Bromiley]

You should google event study stata. There are several programs to do this. If you have the dates of the events, then it is easy to create a variable that equals the return on that date.

[Toby Jones]

You should google event study stata. There are several programs to do this. If you have the dates of the events, then it is easy to create a variable that equals the return on that date.

Hi Phil, thanks for your reply. I had looked into the estudy program, but I hadn't found any evidence that I can use multiple dates. Is this possible?

Thanks

[Toby Jones]

You should google event study stata. There are several programs to do this. If you have the dates of the events, then it is easy to create a variable that equals the return on that date.

As an FYI I have had to start a new thread concerning this issue: https://www.statalist.org/forums/for...multiple-dates.