Dear Ruth,<br><br>now that I changed the specification as 'ego' effect I could start the model, but R then crashed.<br><br>This is the command I used and the output I got after the crash:<br><br>> myeff<-includeInteraction(myeff,egoX,outActSqrt,interaction1="dev.ten",include=TRUE)<br>
name shortName type interaction1<br>collaboration.objEffects.138.1 collaboration unspInt eval <br> interaction2 include effect1 effect2 effect3<br>
collaboration.objEffects.138.1 TRUE 24 17 0<br><br>> ans <- siena07(mymodel, data=mydata, effects=myeff, batch=TRUE, verbose=TRUE, nbrNodes=3, useCluster=TRUE, initC=TRUE)<br><br>Stochastic approximation algorithm.<br>
Initial value for gain parameter = 0.3464102.<br>Start of the algorithm.<br>Target function values are <br> 1. 904.00000 2. 89.00000 3. 1009.38249 4. 5637.09646 5. -23.27556 <br> 6. 354.12048 7. 72.53726 8. 221.19411 9. -16.00520 10. 2756.24970 <br>
11. 2756.24970<br><br>Start phase 0 <br>theta: -1.55 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 <br>Current parameter values:<br>-1.547494 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 <br>
0.000000 0.000000 0.000000<br><br>Start phase 1 <br>Error in unserialize(node$con) : error reading from connection<br><br><br>Best regards,<br><br>Marco<br><br><div class="gmail_quote">On Tue, Mar 16, 2010 at 11:28 PM, Ruth M. Ripley <span dir="ltr"><<a href="mailto:ruth@stats.ox.ac.uk">ruth@stats.ox.ac.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Dear Marco,<br>
<br>
Sorry, I think the information I have coded in the package is incorrect. I<br>
think the egoX should be an 'ego' effect not a 'dyadic': if you change<br>
this in your effects object it should be OK.<br>
<br>
I will fix it in the package but it will be a couple of days before it is<br>
available in the binary on R-forge.<br>
<br>
Regards,<br>
<br>
Ruth<br>
<font color="#888888">--<br>
Ruth M. Ripley, <a href="mailto:Email%3Aruth@stats.ox.ac.uk">Email:ruth@stats.ox.ac.uk</a><br>
Dept. of Statistics, <a href="http://www.stats.ox.ac.uk/%7Eruth/" target="_blank">http://www.stats.ox.ac.uk/~ruth/</a><br>
University of Oxford, Tel: 01865 282851<br>
1 South Parks Road, Oxford OX1 3TG, UK Fax: 01865 272595<br>
</font><div><div></div><div class="h5"><br>
<br>
On Tue, 16 Mar 2010, marco tonellato wrote:<br>
<br>
> Dear All,<br>
><br>
> I'm currently modeling a bipartite network of 83 x 577 nodes with RSiena<br>
> (revision 0.61). I was able to estimate significant and convergent<br>
> parameters for the endogenous part and the main effects of 5 attributes<br>
> (related to either node set).<br>
><br>
> I now have a problem including interaction effects. In particular I would<br>
> like to include the interaction of a constant binary covariate (which<br>
> captures whether or not my nodes of the first mode have tenure) with the<br>
> outdegree activity.<br>
><br>
> Both using the tcl/tk initerface and entering (also with RSienaTest r62):<br>
><br>
> myeff<-includeInteraction(myeff,egoX,outActSqrt,interaction1="dev.ten",include=TRUE)<br>
><br>
> R returned this error message:<br>
><br>
> invalid interaction specification:"must be at least one ego or both dyadic<br>
> effects".<br>
><br>
> Finally, I also noticed that the "interactionType" attribute of the tenure<br>
> egoX effect is set as dyadic, whereas it is left unspecified for all the<br>
> altX effects related to the other attributes I measured for the nodes of the<br>
> second mode.<br>
><br>
> Am I doing something wrong? What should I do to get the interaction effects<br>
> working?<br>
> Thanks a lot,<br>
><br>
> Marco<br>
><br>
<br>
</div></div></blockquote></div><br>