Re: st: mmregress question

[bcoric@efst.hr]

Than you Ronan.

Actually I am getting different results with the same command, mmregress
(as stressed at the p. 444. command is used by implementing sregress and
mmregress, respectively) results with command rreg are always the same.

I agree that proper detection of process which create outliers ,and its
modelling would help to solve the problem. However, the problem is that I
cannot properly detect outliers because outliers detected by
implementation of lvr2plot differ from the outliers which are detected by
mmregress. Furthermore, mmregress does not just produce different results
in each case, but also detects different outliers.

Could you please tell me authors and title of the STB article, you suggest.

Bruno



On 2011 Noll 1, at 17:59, <bcoric@efst.hr> wrote:
>
>> I would like to ask for help with implementation of stata command
>> mmregress.
>> I am doing simple cross section analysis. Since I have just 48
>> observations I am very much concerned about the possible influence of
>> outliers.
>> Previously I was using stata command rreg for such chases. However, I
>> came
>> across the paper, Robust regression in Stata (2009), which argues that
>> rreg command does not have expected robustness properties and recommend
>> mmregress instead.
>> However, I faced some problems in its implementation. Namely, its
>> subsequent implementation to the same model leads to different values of
>> regression coefficients. Moreover, detected outliers also change.
>> I presume, that algorithm use iterative procedure taking previous
>> estimates as starting values. However, results are very different in
>> respect to the sign, size and significance of coefficients, and do not
>> converge after subsequent applications.
>> Does anyone can tell me is there anything, some procedure that I should
>> follow, to get robust (consistent) results.
>
> I would strongly recommend that you read the associated STB article on the
> potential problems with the various methods of robust regression that you
> have used. The fact that you are getting substantively different models
> with different methods points to significant data problems, and in the end
> you may just have to go with the most theoretically justifiable model.
>
> Note also that there are other ways of dealing with outliers, depending on
> the processes that you suspect is at work in producing them. There's a lot
> to be said for trying to develop a model that includes these processes.
>
> Ronán Conroy
> rconroy@rcsi.ie
> Associate Professor
> Division of Population Health Sciences
> Royal College of Surgeons in Ireland
> Beaux Lane House
> Dublin 2
>
>
> *
> *   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/
>


*
*   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/