[Limdep Nlogit List] correcting standard errors for clustered survey data

[Peter Glick]

Bill,

Thanks for elaborating.

Peter

At 11:21 PM 6/22/2006 -0400, you wrote:
>Peter.
>The NLOGIT command does not respect ;Cluster.  It will be added in version 9,
>which is not out yet.  As for my skepticism, the reason is that I want to 
>see the
>formal model for which it gives the right answer.  It is always justified 
>by some
>intuitive appeal - the observations are clustered, so we should do something.
>I do not recall seeing a formal derivation in which it is the appropraite 
>estimator
>for the covariance matrix.  I don't deny that it might be right.  It may 
>well be. I want
>to see the proof.  Stata has so successfully marketed this idea that it 
>has  become
>something people do all the time even tho (based on my many experiences),
>they often have no idea why, or even what it is.
>/Bill Greene
>
>************************************************
>Professor William Greene
>Department of Economics
>Stern School of Business
>New York University
>44 West 4th St., Rm. 7-78
>New York, NY   10012
>Ph. 212.998.0876
>Fax. 212.995.4218
>URL. http://www.stern.nyu.edu/~wgreene
>Email. wgreene at stern.nyu.edu
>************************************************
>
>----- Original Message -----
>From: Peter Glick <pjg4 at cornell.edu>
>Date: Thursday, June 22, 2006 3:00 pm
>Subject: Re: [Limdep Nlogit List] correcting standard errors 
>for        clustered survey data
>
> > Bill,
> >
> > Thank you for your reply.
> >
> > Unfortunately, while the cluster command seems to work fine for
> > simpler
> > models (provided the data are first sorted by cluster) it does not
> > for
> > nlogit.  I am using the setup where the data are arranged one line
> > per
> > alternative and the utility fns are written in the model
> > statement.  Essentially, nlogit does not carry out the cluster
> > command,
> > though it clearly reads it since putting in an incorrect cluster
> > name leads
> > to an error message.
> >
> > Is there anything I can do? Has the option been extended to nlogit
> > in an
> > updated version? We have nlogit v8.
> >
> > An another note, your skepticism about the cluster estimator has
> > me curious
> > as I have not heard this except in regards to problems when there
> > is
> > a  small number of clusters (discussed by e.g., Wooldridge and
> > Bertrand,Duflo, and Mullainathan).  Is there a critical paper on
> > this by
> > you or others?
> >
> > Thanks again for your help..
> >
> > Peter
> >
> > At 11:21 AM 6/22/2006 -0400, William Greene wrote:
> > >Peter.
> > >The 'cluster' correction is a different thing from the 'robust'
> > estimator,
> > >I would argue.
> > >The issue is methodological.  My point is the one that you make
> > in your
> > >(1). About
> > >the cluster estimator, while I am a little skeptical of it, I
> > have
> > >observed in applications
> > >that it seems to do what it is argued to be attempting to do.
> > So, I would
> > >agree with
> > >your point 1.  On 2, you should be able to use ;CLUSTER = spec.
> > where spec
> > >is either
> > >a number indicating a fixed cluster size or a variable name that
> > contains
> > >a group
> > >identifier such as the cluster number.
> > >/Bill Greene
> > >
> > >************************************************
> > >Professor William Greene
> > >Department of Economics
> > >Stern School of Business
> > >New York University
> > >44 West 4th St., Rm. 7-78
> > >New York, NY   10012
> > >Ph. 212.998.0876
> > >Fax. 212.995.4218
> > >URL. http://www.stern.nyu.edu/~wgreene
> > >Email. wgreene at stern.nyu.edu
> > >************************************************
> > >
> > >----- Original Message -----
> > >From: Peter Glick <pjg4 at cornell.edu>
> > >Date: Thursday, June 22, 2006 11:02 am
> > >Subject: [Limdep Nlogit List] correcting standard errors for
> > >clustered  survey data
> > >
> > > > Dear Professor Greene (and list members),
> > > >
> > > > I would like adjust my multinomial logit standard errors for
> > > > sample design
> > > > (correlations of errors within clusters).  I was hoping there
> > > > would be a
> > > > general command for this in Limdep (I have version 8)  but there
> > > > seems not
> > > > to be.  Assuming this is correct, it would be very helpful to have
> > > > a few
> > > > points clarified:
> > > >
> > > > 1.   The manual (under 'Robust Standard Errors') warns against
> > > > using the
> > > > Huber White robust covariance matrix for nonlinear models because
> > > > the very
> > > > need for this correction, e.g.,  heteroskedasticity, in effect
> > > > means the
> > > > parameters themselves are not consistent.  I assume that this does
> > > > not
> > > > refer to intra-cluster error correlations (violation of the
> > > > independence
> > > > assumption), since if it did, parameter estimates from just about
> > > > all
> > > > nonlinear models using standard household surveys would be
> > > > inconsistent.  Am I correct?
> > > >
> > > > 2.   Despite the warning, the manual notes that for the discrete
> > > > choice
> > > > models that allow choice based sampling, one can obtain the White
> > > > robust
> > > > standard errors using: "; Wts = ONE ; Choice Based".   But I
> > > > believe this
> > > > would yield the standard sandwich estimator which would not be the
> > > > appropriate one for correcting for cluster correlations of the
> > > > errors, i.e,
> > > > it assumes independence.  Is this correct?   If so, I would need
> > > > to
> > > > construct the appropriate matrix using the scores, etc.  Has
> > > > anyone done
> > > > this and could they provide code?
> > > >
> > > > Many thanks..
> > > >
> > > > Peter Glick
> > > > Cornell University
> > > > _______________________________________________
> > > > Limdep site list
> > > > Limdep at limdep.itls.usyd.edu.au
> > > > http://limdep.itls.usyd.edu.au
> > > >
> > >
> > >_______________________________________________
> > >Limdep site list
> > >Limdep at limdep.itls.usyd.edu.au
> > >http://limdep.itls.usyd.edu.au
> >
> > _______________________________________________
> > Limdep site list
> > Limdep at limdep.itls.usyd.edu.au
> > http://limdep.itls.usyd.edu.au
> >
>
>_______________________________________________
>Limdep site list
>Limdep at limdep.itls.usyd.edu.au
>http://limdep.itls.usyd.edu.au