[Rsiena-help] multiplication factor

[Tom Snijders]

Dear Bethany and Ruth,

Indeed the role of autocorrelations for the Method of Moments is unrelated to the multiplication factor which is used for maximum likelihood estimators (both ERGM and for the longitudinal SAOM) and currently not yet in RSiena. 

The proof of the pudding is in the convergence t-ratios produced in phase 3. If these are small enough, it does not matter if autocorrelations during phase 2 are high. 

Best wishes,

Tom

============================================
Tom A.B. Snijders
Professor of Statistics in the Social Sciences
Department of Politics and Department of Statistics
Nuffield College
University of Oxford 
tel. +44-01865-278599
http://stat.gamma.rug.nl/snijders/

> -----Original Message-----
> From: rsiena-help-bounces at lists.r-forge.r-project.org [mailto:rsiena-help-
> bounces at lists.r-forge.r-project.org] On Behalf Of Ruth M. Ripley
> Sent: 03 March 2010 16:36
> To: bwotal at cyrus.psych.uiuc.edu
> Cc: Rsiena-help at r-forge.wu-wien.ac.at
> Subject: Re: [Rsiena-help] multiplication factor
> 
> Dear Bethany,
> 
> Do you mean the warnings in Phase 2, which appear only if you have the
> verbose option? If so, you could try increasing the number of subphases
> from the default 4. If this does not help, as this is not a problem with
> the software, you'll need help from more experienced users than me. I've
> copied this message to the mailing list, in the hope that someone else who
> reads it may respond!
> 
> Regards,
> 
> Ruth
> --
> Ruth M. Ripley,                         Email:ruth at stats.ox.ac.uk
> Dept. of Statistics,                    http://www.stats.ox.ac.uk/~ruth/
> University of Oxford,                   Tel:   01865 282851
> 1 South Parks Road, Oxford OX1 3TG, UK  Fax:   01865 272595
> 
> On Wed, 3 Mar 2010, bwotal at cyrus.psych.uiuc.edu wrote:
> 
> > Yeah- I am getting some autocorrelations that are too high and the only
> info
> > I could find on that was in Siena3.
> >
> > Is there another way to help solve that error message?
> >
> > Bethany
> >
> >
> > Quoting "Ruth M. Ripley" <ruth at stats.ox.ac.uk>:
> >
> >> Dear Bethany,
> >>
> >> I'm not sure what you are doing. The M-H algorithm is not implemented
> yet.
> >> Are you reading the manual for Siena3?
> >>
> >> Regards,
> >>
> >> Ruth
> >> --
> >> Ruth M. Ripley,                         Email:ruth at stats.ox.ac.uk
> >> Dept. of Statistics,
> http://www.stats.ox.ac.uk/~ruth/
> >> University of Oxford,                   Tel:   01865 282851
> >> 1 South Parks Road, Oxford OX1 3TG, UK  Fax:   01865 272595
> >>
> >>
> >> On Tue, 2 Mar 2010 bwotal at cyrus.psych.uiuc.edu wrote:
> >>
> >>> I'm running into some autocorrelations that are higher than I'd like.
> >>>
> >>> Are there options in the RSiena code to change the multiplication
> >>> factor and improve the performance of the M-H algorithm?
> >>>
> >>> Thanks!! Bethany
> >>>
> >>> _______________________________________________
> >>> Rsiena-help mailing list
> >>> Rsiena-help at lists.r-forge.r-project.org
> >>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rsiena-
> help
> >>>
> >>
> >>
> >>
> >>
> >
> >
> >
> >
> _______________________________________________
> Rsiena-help mailing list
> Rsiena-help at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rsiena-help