Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: autocorrelation and cross-correlation analysis after pooled OLS-regression


From   "Maren Hartmann" <maren.hartmann@gmx.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: autocorrelation and cross-correlation analysis after pooled OLS-regression
Date   Thu, 03 Sep 2009 20:55:22 +0200

Dear Stata-Users,

I am quite new to stata and have run a pooled OLS-regression and am now trying to check for autocorrelation (my datafile consists of stacked time-series) and cross correlation (as "individual" time series overlap) in my residuals.

To check for autocorrelation, I installed -panelauto- but could not find the helpfile or syntax for the command - thus I don't even know if it is the right thing to do. Does anyone know how I can best check for autocorrelation in the residuals of my pooled OLS-Regression?
Is there any way to output the results so I can further analyse them?

To check for cross-correlation I reshaped and merged the data so that I have the "individual time series residuals" along one single timeline and used the -mkcorr- command - I would prefer to use the -pwcorr- command (as with pwcorr the number of observations used are included) but as I understand, it is not possible to output the results obtained by pwcorr. Is there maybe another way to efficiently correlate the pairwise correlations and output the data?

Also, -mkcorr- returned an error code 2000 - no observations - how do i interpret this with the -mkcorr- command and is there any way around this?

Many thanks in advance for any ideas concerning either of the above questions

Kind regards
Maren
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index