st: AW: atrho0 option

["Meulemann Max" <mmeulemann@ethz.ch>]

So just in case anyone else has that question again.

Defining the matrix is defining a vector
So any 
matrix V=(rho21, rho31,...) up to M(M-1)/2 

And then properly define the colnames
matrix colnames V=atrho21:_cons atrho_31_cons               (no brackets apparently) 
According to the rhos given in V for which you would like to predefine values.

So if I have 3 equation and I would like to predefine rho21 and rho32, I write
matrix V=(0.3, 0.2) 
matrix colnames V=atrho21_cons atrho32:_cons
mvprobit (y1 x) (y2 x), atrho0(V)

-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Meulemann Max
Gesendet: Dienstag, 13. August 2013 16:39
An: statalist@hsphsun2.harvard.edu
Betreff: st: atrho0 option 

Im using stata 12 and would like to use the atrho0() option of mvprobit

Im having a hard time defining the matrix for this and I really cannot find any example although there have been thread on this here before.

Say I have 3 equations I want to estimate with mvprobit mvprobit (y1 x) (y2 x) (y3 x)

How do I predefine the right matrix and the matrix colnames command to estimate

mvprobit (y1 x) (y2 x) (y3 x), atrho0(test)

do i define?
matrix test= ( 1 ,r21, r31 \ r21, 1, r32 \ r31, r32, 1)

what do I do for the matrix colnames ?

Best
Max
-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Max
Gesendet: Dienstag, 13. August 2013 15:58
An: statalist
Betreff: st: -estout- -- Adding significance stars for user-computed statistics

Hello all,

I'm using -estout- to produce a latex table of regression results and I ran into a question about adding significance stars to statistics that I added using -estadd-. In a regression, I calculate a statistic for several independent variables (lets call the independent variables A, B, and C). When I run the regression and store the results I also use estadd in the following way:

foreach x in A B C {
  estadd scalar pct_shift_`x' = 100*_b[`x']/r(mean) }

So I add three scalars for each regression (pct_shift_A, pct_shift_B, pct_shift_C). When outputting the latex table, I do:
estout using output.tex,
cells(none) stats(pct_shift_A pct_shift_B pct_shift_C)

This produces a table with the percent shifts. What I would like is to have the significance stars that are associated with A, B, and C be attached to the added statistics pct_shift_A, pct_shift_B, and pct_shift_C. But I can't figure out how to go about doing that. Any advice would be greatly appreciated.

Thanks,
Max

*
*   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/