[Rsiena-help] coding missings

[Ruth M. Ripley]

Dear Chip,

Maybe your networks are data frames not matrices. (try 
class(friend.data.w1) to find out, and 
friend.data.w1 <- as.matrix(friend.data.w1)
to change it.

Ask again if this is not the problem.

Regards,

Ruth


On Wed, 16 Jun 2010, Huisman, Chip wrote:

> Dear all,
>
> I get the following error message:
>
>
> Error in sienaNet(array(c(friend.data.w1, friend.data.w2), dim = c(962,  :
>  entries in networks must be 0, 1, 10 or 11
>
> I use 9 for missings and 10 for structural sero's.
>
> and I use the following script to indicate the missings:
> friend.data.w1[friend.data.w1 %in% c(9)] <- NA
> friend.data.w2[friend.data.w2 %in% c(9)] <- NA
>
> what am I doing wrong here?
>
> best wishes,
>
> Chip
>
>
>
>