informR (version 1.0-5)

slbind.cov: Add Covariates to a Statslist Array

Description

Combines scalars and vectors to statslist arrays using a wrapper for abind() in the abind package.

Usage

slbind.cov(covar, statslist, type = 1,...)

Arguments

covar

A list of lists containing the covariates, see details

statslist

A statslist object, possibly passed from gen.intercepts

type

Indicates where the combining is going to occur in statslist. 1 for global, 2 for local.

...

Additional methods passed to abind().

Details

The length AND actor-order of the covar list must equal that of the statslist. Currently, no check is made to ensure that the actor-order is maintained, though an object length error will be returned in many faulty cases. Also, note that the number of covariates in each element of covar (i.e., for each actor) should be the same for proper model identification (also not currently checked).

Actor-level covariates are passed as single values and event-level covariates are passed as vectors. That is, each sub-element (covariate) of the covar list must either be a scalar or a vector with length equal to length(eventlist$eventlist$actor).

If names(covar$actor) (for any and each actor) is NULL, then slbind.cov() will generate names using the make.names=TRUE parameter, as discussed in the ?abind documentation.

See Also

slbind, abind, slbind.cond

Examples

rawevents<-sample(rep(c("ran","eat","stay","eat","ran","play"),50))
actors<-rep(c("Jim","Bill","Pete"),100)
evmat<-cbind(rawevents,actors)
eventlist<-gen.evl(evmat)
beta.ints<-gen.intercepts(eventlist)

##Make up some covariates
covar<-list()
covar$Bill$rate<-sort(rexp(length(eventlist$eventlist$Bill)))
covar$Bill$smokes<-1
covar$Jim$rate<-sort(rexp(length(eventlist$eventlist$Jim)))
covar$Jim$smokes<-0
covar$Pete$rate<-sort(rexp(length(eventlist$eventlist$Pete)))
covar$Pete$smokes<-0

statslist.new<-slbind.cov(covar,beta.ints)
statslist.new[[1]][[1]][,1,]