Pooled Cross Section

[Neudelinho]

Hi everybody!
I write my master thesis about the win/loss effect of soccer clubs on their stock price. I have nice results, but now I want to summarize them to a general statement.

I have 13 clubs which are listed on different dates. Ajax data is from 1.1.1999, Lyon from 09.02.2007, Trabzonspor from 18.04.2005 and so on.

For my regression I use Eviews 7 and an excel sheet with the columns:
Date / Abnormal return / Win (dummy v) / Draw (dummy) / Loss (dummy)

My friend said that I should use dummy variables for years. So I created some for 1999, 2000,..., 2013.
Now I regressed abnormal returns on win / draw / loss /1999...2013 and include a constant. Not included is the date and the resid (which Eviews created on its own). If I run the equation, EViews says "near singular matrix error. Regressors may be perfectly collinear." This might be the reason that there are dates which appears often. As I have 13 teams there is for example 13 times the date with 27.02.2012, 04.01.1999 is there 4 times listed.

So my question is whether Eviews is able to solve my problem or should I use excel? As my econometrics knowledge is not that good, please give me a hint how to solve my problem. My whole work is based on generilizing the results of 13 teams to one result. Please help me. I want to thank you in advance!

[Newbie12]

Hi there,

Sorry I'm not actually answering your question but I think you could probably help me.
I have two companies Cost = Quantity(subscript)t + Quantity^2(subscript)t + Quantity^3(subscript)t + Error(subscript)t
Cost(subscript)2 = Quantity(subscript)t + Quantity^2(subscript)t + Quantity^3(subscript)t + Error(subscript)t
How do I pool these two companies data together ?
Thanks, sorry for it being so basic.