Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: RE: RE: creating AR(1) correlation matrix

 From Federico Belotti To statalist@hsphsun2.harvard.edu Subject Re: st: RE: RE: creating AR(1) correlation matrix Date Thu, 16 May 2013 18:17:54 +0200

```You can easily include the mata code into a do file, convert the mata matrix to Stata and use it as an argument of -drawnorm-

local n 1000
local rho 0.9
local dim 10
mata: rho = `rho'
mata: dim = `dim'
mata: Lambda = I(dim)
mata: Lambda =  rho:^abs(J(1,dim,(1::dim))-J(dim,1,(1..dim)))
mata: st_matrix("Lambda", Lambda)
local xs
forvalues i=1/`dim' {
local xs "`xs' x`i'"
}
clear
set obs `n'
drawnorm `xs', corr(Lambda) n(`n')
corr x*
sum x*

HTH
Federico

On May 16, 2013, at 6:01 PM, Lachenbruch, Peter wrote:

> I need to generate the correlation matrix for input to drawnorm, so this isn't quite what i need.  Thanks for the help
>
> Peter A. Lachenbruch,
> Professor (retired)
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Jacobs, David [jacobs.184@sociology.osu.edu]
> Sent: Thursday, May 16, 2013 8:13 AM
> To: 'statalist@hsphsun2.harvard.edu'
> Subject: st: RE: creating AR(1) correlation matrix
>
> A quick and maybe dirty way to do this is to run a population averaged, panel regression model with an AR1 term if that's appropriate.
>
> Then type -xtcorr- (there's a newer command but xtcorr still works at least in version 11).  You'll probably get the correlation matrix used to compute the AR1 correction.  I say probably because the matrix you want may be too large for this option.
>
> Dave Jacobs
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Lachenbruch, Peter
> Sent: Thursday, May 16, 2013 10:59 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: creating AR(1) correlation matrix
>
> I need to create a correlatoin matrix of rho^(i-j) where i and j are row and column indexes.  Is there a simple command in Stata that will do htis?  I can do it in a foreach loop (although i've been messing it up so far).  This is for a simulation and the dimension of the correlation is 50 or 100 so i don't want to do it manually.
>
>
> Peter A. Lachenbruch,
> Professor (retired)
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

--
Federico Belotti, PhD
Research Fellow
Centre for Economics and International Studies
University of Rome Tor Vergata
tel/fax: +39 06 7259 5627
e-mail: federico.belotti@uniroma2.it
web: http://www.econometrics.it

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```