Bookmark and Share

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


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

Re: st: Odd results with -estimates- command


From   [email protected] (Jeff Pitblado, StataCorp LP)
To   [email protected]
Subject   Re: st: Odd results with -estimates- command
Date   Mon, 04 Nov 2013 13:27:56 -0600

Eric DE SOUZA <[email protected]> using -estimates table- to report a
table of common coefficients in -regress-, -logit-, and -probit- model fits
but not all the requested coefficients are being reported:

> The output below explains the subject header.
> If bOLS is placed in the last column it does not print
> If bOLS is placed in the first column, the other two do not print.
> 
> . use ch13.dta
> 
> . reg fishmode lnrelp
> 
> . estimates store bOLS
> 
> . logit fishmode lnrelp
> 
> . estimates store blogit
> 
> . probit fishmode lnrelp, nolog
> 
> . estimates store bprobit
> 
> . estimates table blogit bprobit bOLS, t stats(N ll r2 r2_p) b(%10.4f) keep(_cons lnrelp)
> 
> -----------------------------------------------------
>     Variable |   blogit      bprobit        bOLS     
> -------------+---------------------------------------
>        _cons |     2.0531       1.1944               
>              |      12.15        13.34               
>       lnrelp |    -1.8225      -1.0555               
>              |     -12.61       -13.87               
> -------------+---------------------------------------
>            N |        630          630          630  
>           ll |  -206.8270    -204.4109    -195.1673  
>           r2 |                               0.4633  
>         r2_p |     0.4486       0.4550               
> -----------------------------------------------------
>                                           legend: b/t
> 
> . estimates table bOLS blogit bprobit, t stats(N ll r2 r2_p) b(%10.4f) keep(_cons lnrelp)
> 
> -----------------------------------------------------
>     Variable |    bOLS        blogit      bprobit    
> -------------+---------------------------------------
>        _cons |     0.7842                            
>              |      58.21                            
>       lnrelp |    -0.2429                            
>              |     -23.28                            
> -------------+---------------------------------------
>            N |        630          630          630  
>           ll |  -195.1673    -206.8270    -204.4109  
>           r2 |     0.4633                            
>         r2_p |                  0.4486       0.4550  
> -----------------------------------------------------
>                                           legend: b/t

We suspect that the problem is due to -logit- and -probit- attaching an
equation name for the coefficients in -e(b)-, but -regress- is not.

A work around is to replace the -keep(_cons turn)- option with -eq(1)-.

--Jeff
[email protected]
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index