RE: st: Standard deviation of the error term in asclogit

[Cameron McIntosh <cnm100@hotmail.com>]

Hi Ester,
It's my impression that these quantities are not estimable because the  
error terms in probit and logit models are fixed at the variances of  
these distributions (1.0 and 3.29), respectively, for identification  
purposes. However, unobserved heterogeneity can still cause us  
problems, and there are some work-arounds that attempt to identify and  
quantify the degree of heterogeneity: 

Fielding, A. (2004). Scaling for residual variance components of  
ordered category responses in generalised linear mixed multilevel  
models. Quality and Quantity, 38, 425-433. 

Bauer, D.J. (2009). A note on comparing the estimates of models for  
cluster-correlated or longitudinal data with binary or ordinal  
outcomes. Psychometrika, 74(1), 97-105. 

Allison, P.D. (1999). Comparing logit and probit coefficients across  
groups. Sociological Methods & Research, 28(2), 186-208. 

Mood, C. (2010). Logistic regression: Why we cannot do what we think we  
can do, and what we can do about it. European Sociological Review,  
26(1), 67-82. 

Williams, R. (2009). Using heterogeneous choice models to compare logit  
and probit coefficients across groups. Sociological Methods & Research,  
37(4), 531-559. 

Long, J.S. (2009). Group Comparisons in Logit and Probit Using  
Predicted Probabilities. Working paper draft 2009-06-25. 
http://www.indiana.edu/~jslsoc/files_research/groupdif/groupwithprobabilities/groups-with-prob-2009-06-25.pdf http://www.indiana.edu/~jslsoc/research_groupdif.htm 

Sorry if I am misunderstanding your situation... others will probably  
chime in with additional comments and suggestions. 

Cam 
>  
>  > Date: Thu, 8 Sep 2011 14:04:01 +0100 
>  > From: ester_adler@yahoo.com 
>  > Subject: st: Standard deviation of the error term in asclogit 
>  > To: statalist@hsphsun2.harvard.edu 
>  > 
>  > I have a dataset whereby each of a number of individuals are  
> presented with a number of menus, each menu consisting of a number of  
> options, of which they must choose one. The same set of menus is  
> presented to each individuals. I run asclogit (as well as asmprobit and  
> mixlogit) on the data. 
>  > 
>  > My issue is that I have reasons to believe my individuals have  
> different standard errors for their error term; by this I mean that  
> some are more "accurate" than others in their choice. Is there a way I  
> can get Stata to take account of this? 
>  > 
>  > Also, since I have variables that I suppose may explain how accurate  
> the individual is in her choices (for example, mathematical proficiency  
> and education), is there a way to get the std error or the error term  
> to depend on this? 
>  > 
>  > Thanks! 
>  > 
>  > Ester 
>  > 
>  > * 
>  > * 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/