[Rsiena-help] clarification on changelog for RSienaTest rev.93

[marco tonellato]

Dear Ruth and Tom,

thanks for the clarifications.

By the way, since we don't have missing data but just a composition change
file, then this is probably why we didn't experience the estimation to be
particularly slow.

The print01 correctly reported the actors for both modes leaving and joining
the network at times specified in the composition change. Is this
information currently used during the estimation?

thanks again,
regards,

Marco

On Tue, Jun 8, 2010 at 11:34 AM, Tom Snijders <
Tom.Snijders at nuffield.ox.ac.uk> wrote:

> Dear Ruth,
>
> Thank you for your answer. Given other things, I do not think this should
> be given high priority. My message was mainly intended to notify Marco (and
> perhaps others) that the functionality of bipartite networks in RSiena is
> still incomplete, and that anybody using it should be aware that this just
> is something that we are still working on.
>
> Best,
>
> 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: Ruth M. Ripley [mailto:ruth at stats.ox.ac.uk]
> > Sent: 08 June 2010 11:32
> > To: Tom Snijders
> > Cc: 'marco tonellato'; rsiena-help at lists.r-forge.r-project.org
> > Subject: Re: [Rsiena-help] clarification on changelog for RSienaTest
> > rev.93
> >
> > Dear Tom,
> >
> > No, the actors in the ministeps do not yet have an option to leave things
> > unchanged. Implementation of this is non trivial, compared to the other
> > problems (C++ not R!), so I cannot promise it will be done very soon.
> >
> > But I will see what I can do.
> >
> > Regards,
> >
> > Ruth
> >
> > On Tue, 8 Jun 2010, Tom Snijders wrote:
> >
> > > Dear Marco,
> > >
> > > Note that the bipartite network version of RSiena still is a beta
> > > version, and it is not yet totally clear to me that it does what it
> > > should do eventually. In particular, I am not sure that in the
> > > ministeps, the option to leave things unchanged is implemented yet.
> > > Ruth, perhaps you could also comment on whether this option is
> > > implemented?
> > >
> > > 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/
> > >
> > > ________________________________
> > > 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 marco tonellato
> > > Sent: 08 June 2010 11:14
> > > To: rsiena-help at lists.r-forge.r-project.org
> > > Subject: [Rsiena-help] clarification on changelog for RSienaTest rev.93
> > >
> > > Dear Ruth,
> > >
> > > I just read the changelog for RSienaTest rev.93 (04.06.2010), and I saw
> > those messages:
> > > "Using composition change with bipartite networks will give an error
> > message - until this is corrected". And "Bug fix to treatment of missing
> > data in sparse format bipartite networks".
> > >
> > > Since I'm actually using sparse bipartite networks with composition
> > change (RSiena rev.92) I was wondering whether the problem was there
> > before and you fixed it or whether the problem wasn't there before and
> you
> > signaled it.
> > > Can you please elaborate on that?
> > >
> > > Thanks. Regards,
> > > Marco
> > >
> > >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rsiena-help/attachments/20100608/e8b5e7f0/attachment-0001.htm>