urgent please- adjusted R-square after VAR model?

[abdelrazzaq alrababaa]

Can anyone please advice how to get the adjusted r-square after the VAR model. I am adding more exogenous variables, so I need to see the adjusted r-square?

[Nick Cox]

On claims of urgency, please see the comments within http://www.statalist.org/forums/foru...riables-urgent

Otherwise show your commands and results to give a clear indication of exactly what you are doing. Which literature leads you to suppose that adjusted R-square is a good idea for VAR?

[abdelrazzaq alrababaa]

Thank you Nick;
my command
var djiaorg djtrasnorg nas100org nasbioorg nasfinorg SPsmallorg, lags(1/1) exog(cpi ppindex export import persincome indproduc caputi p
mi concredit unemp AvgHrlyEarnings factorygood businventory retail housingstarts newhousesales retail) lutstats
it is standard VAR with stock indexes return as endogenous and macro economic surprises as exogenous. the idea is to start with stock indexes return and add exogenous to investigate the role of macro announcements in spillover- the model can be considered as a forecasting encompassing regression with more macro economic factors to be added, hence it is important to compare the performance of the models among the sectoral indexes with and without the macro announcements, adding more insignificant exogenous factors will definitely affect the r2 value and would be better use the adjusted r square instead. I will start with the full specification model with all announcements and then including the significant announcements only.

[Nick Cox]

Thanks for the extra detail. I see no results here, or literature references, just the commands you typed and more on what you are trying to do.

My aim was to get you to improve your question so that people expert in VAR could comment. I have never used it, so I sign off here.

[abdelrazzaq alrababaa]

Thank you. I added the link below for the most related article.

[Friedrich Huebler]

Nick gave good advice. Please read the FAQ, especially sections 12 and 13. I am sorry but the link you provided is not helpful.

[abdelrazzaq alrababaa]

Dear Friedrich Huebler;

below are the attached files, my question and clarification, related paper.
also please check out the following link :

http://www.haghish.com/statistics/st...oad/r2var.html

skip most of contents and go to the end of the page . the author was able to report the adjusted r square some formulas as shown in the top of the page, although I am unable to use these formulas in Stata as they seem old and my Stata 13 version does not accept them. Could you please advice.

[Friedrich Huebler]

I am sorry to insist, but you have clearly not read the FAQ. Here is a quote from section 12:

You can attach datasets or other documents, but that is usually much less convenient than the methods above. Note, in particular, that MS Word and MS Excel file formats are not universally readable by forum members.

You are also not following Nick's advice.

...show your commands and results to give a clear indication of exactly what you are doing.

[abdelrazzaq alrababaa]

I am sorry. my command and results attached now again in the new .txt file

[abdelrazzaq alrababaa]

Dear Friedrich Huebler;
I suppose that my question is clear now.

[Friedrich Huebler]

I regret to inform you that I have nothing to add to what I wrote before. It is in your interest to read the FAQ and act upon the advice given there.

[abdelrazzaq alrababaa]

Many thanks by the way. I don't know what I should do more. I prefer to find the answer elsewhere. should not be that strict. all the information are included!

[Anders Alexandersson]

Try the user-written program -r2var-. The helpfile of -r2var- states:

Adjusted R2 = 1-(1-R2)*((QN-Q)/(QN-K))

where
N = number of observations
K = Number of Parameters
Q = Number of Equations

net describe r2var, from(http://fmwww.bc.edu/RePEc/bocode/r)