[Rsiena-help] Help with includeInteraction()

[Ruth M. Ripley]

Dear Guido,

Sorry, this is a bug. It needs to check whether you are trying to set or
unset an interaction.

I will fix it as soon as I have time. But that won't be for a week or so.

You may be able to unset it using the setEffect function.

Regards,

Ruth


On Fri, 21 May 2010, Guido Conaldi wrote:

> Dear Ruth,
>
> I am trying to include interaction effects in my model and I noticed
> that whereas includeInteraction() do include interaction terms in the
> model, I seem to be unable to exclude them once specified by using the
> include=FALSE option.
>
> For example, I give:
>
> > foo.effects<-includeInteraction(foo.effects, egoX, outActSqrt, interaction1="dev.ten")
>
>
> name shortName type interaction1 interaction2
> foo.coll.objEffects.138.2 foo.coll   unspInt eval
>                           include effect1 effect2 effect3
> foo.coll.objEffects.138.2    TRUE      21      16       0
>
>
> and then:
>
>
> > foo.effects<-includeInteraction(foo.effects, egoX, outActSqrt, interaction1="dev.ten" , include=FALSE)
>
> name shortName type interaction1 interaction2
> foo.coll.objEffects.138.3 foo.coll   unspInt eval
>                           include effect1 effect2 effect3
> foo.coll.objEffects.138.3   FALSE      21      16       0
>
>
> and the list of effects still include the interaction:
>
>
> > foo.effects
>    effectName                        include fix   test  initialValue parm
> 1  constant foo.coll rate (period 1) TRUE    FALSE FALSE    0.97384   0
> 2  constant foo.coll rate (period 2) TRUE    FALSE FALSE    1.73322   0
> 3  outdegree (density)               TRUE    FALSE FALSE   -1.39237   0
> 4  4-cycles                          TRUE    FALSE FALSE    0.00000   2
> 5  indegree - popularity (sqrt)      TRUE    FALSE FALSE    0.00000   0
> 6  outdegree - activity (sqrt)       TRUE    FALSE FALSE    0.00000   0
> 7  bug.ten alter                     TRUE    FALSE FALSE    0.00000   0
> 8  bug.com alter                     TRUE    FALSE FALSE    0.00000   0
> 9  bug.sev alter                     TRUE    FALSE FALSE    0.00000   0
> 10 unspecified interaction effect    TRUE    FALSE FALSE    0.00000   0
> 11 unspecified interaction effect    TRUE    FALSE FALSE    0.00000   0
>    effect1 effect2
> 1   0       0
> 2   0       0
> 3   0       0
> 4   0       0
> 5   0       0
> 6   0       0
> 7   0       0
> 8   0       0
> 9   0       0
> 10 21      16
> 11 21      16
>
> I noticed that the first command in the example sets the effect "foo.coll.objEffects.138.2", whereas the second one sets the effect "foo.coll.objEffects.138.3"
>
> Am I doing something wrong here?
>
> Thank you in advance,
> Regards,
>
> Guido
>
> _______________________________________________
> 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
>