Re: st: RE: difference in number of observation in stata and limdep tobit, probit result

[Nick Cox <njcoxstata@gmail.com>]

This tells us that you have missing values on at least one variable,  
so I think the only guess possible is that this underlies apparent  
differences. Otherwise there is no evidence here that Stata and Limdep  
give different answers to the same question.

Nick

On 5 Aug 2012, at 05:43, Prakash Singh <prakashbhu@gmail.com> wrote:

> Nick, here is the situation
> I have data with total of 15469 observation now when I am running
> probit or tobit is stata and limdep the number of observation in the
> output table is coming to be quite different. The number of
> observation in limdep is 15469 and in stata it is 5709. Below I am
> providing the summary statistics of the variable used in regression.
>
>
>
>  Variable |              Obs        Mean    Std. Dev.        
> Min        Max
> -------------+--------------------------------------------------------
>          FA          |     15469    .2967225    .4568279           
> 0          1
>       SALES      |     15469     4459518    2.47e+07        120    
> 1.28e+09
> prop_owned~t |      6017    .3051433    .2953013   1.25e-06          1
>    dacount2     |     15420    .6050584     .488854           
> 0          1
>  dtype_ent1    |     15469    .0637404    .2442978           
> 0          1
> -------------+--------------------------------------------------------
>  dtype_ent2    |     15469    .4825134    .4997103           
> 0          1
> dum_owner_n2 |     15345    .1571848    .3639867          0          1
> dum_owner_n3 |     15345    .8165526    .3870455          0          1
>       dreg2          |     15449    .3852677    .4866742           
> 0          1
>
>
>
> Hope I have given sufficient information for getting my query clear.


*
*   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/