Writing to a file using esttab r(coefs, transpose)

[Claudia Zapata]

I am running several sets of regressions (100plus sets of 64 regressions) that need to be displayed in a table so that the models are in rows rather than columns. While I am able to do this using esttab, I am not able to write the table produced by esttab to any external files using the customary "using" command. My full code is below. I receive an error matrix r(coefs) not found) when I run the line esttab r(coefs, transpose) using filename,replace. There is no error if I do not use "using".

Is there a conflict between calling the coefficients from Stata and also writing to an external file ? If esttab is not the correct approach, I would be grateful for some advice pointing me in the right direction.

sysuse auto,clear
eststo clear //Clear previously stored regressions.


levelsof rep78, local(levelrep78)
di "`levelsrep78'"

foreach l of local levelrep78 {
di "rep78: `l'"
capture noisily eststo Route`l': reg p turn head if rep78==`l'

}

esttab using r, p scalars(F df_m df_r) keep( turn ) replace
mat list r(coefs)


esttab r(coefs, transpose) using filename,replace //Final Table

matrix r(coefs) not found

Best,

Claudia

[Nils Enevoldsen]

This modification of your code works:

sysuse auto, clear
eststo clear //Clear previously stored regressions.
matrix drop _all

levelsof rep78, local(levelrep78)
di "`levelsrep78'"

foreach l of local levelrep78 {
    di "rep78: `l'"
    capturenoisilyeststo Route`l': reg p turn head if rep78==`l'
}

esttab using r, p scalars(F df_m df_r) keep( turn ) replace 
mat list r(coefs) 
mat rename r(coefs) foo
mat list foo

esttab matrix(foo, transpose) using filename, replace //Final Table

The error is not related to the using option; the error occurs even without it. Rather, it's something about r(coefs) being a special matrix to estout. It is, after all, a matrix created by estout. I sidestepped the issued by renaming the matrix r(coefs) to foo before esttabbing it.

[Claudia Zapata]

Great! Thank you very much for your help. This worked perfectly!