Problem with performing right regression with my data

[xander meertens]

Hello,
I am analyzing the correlation between market cap and abnormal returns of targeted M&A firms, 10 days before the announcement date. My time frame is t-10 untill t-1. My control time frame is t-90 untill t-30.
I currently made this regression code.
reg Cumulated_abn_returns MarketCap TargetDebtRatio BTMValue Cash Liquidity yeardummy's. Now I heard that I'm dealing with panel data, so I performed a xtreg.
This is the regression used for this problem. xtreg CAR_EVENT Liquidity BTMValue Cash TargetDebtRatio dummy_2005 dummy_2004 dummy_2003 dummy_2006 Same_Industry Firmvalue. Beforeof the regression I used xtset ID, T_Event.
I also posted the link to my dataset. Hopefully you can help us solve this problem.

Is this the right way to perform this regression test? I also want to test the correlation of liquidity and MarketCap with a similar regression. Is that possible?

Kind regards,
Xander Meertens



Attached Files

• 

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Pictures and files are not recommended - many of us won't open files from someone we don't know.

I can't judge the entire set of your work, but the regression is fine. It suggests a substantial (.14) increase in CAR for a one unit increase in liquidity. Since liquidity is often fraction, you won't have a one unit change in the data so the large size of the coefficient is probably OK. You can replace CAR with market cap if you want and run the model. However, firmvalue might be automatically associated with market cap (depending on how you measure firm value).