bcov {hglmm}R Documentation

Bootstrap covariance.

Description

Creates an estimate of the covariance matrix of the parameter estimates for a hidden generalized linear Markov model via parametric bootstrapping.

Usage

   bcov(object, nsim = 500, itmax = 500, verbose = FALSE,
        data.name = NULL)

Arguments

object

An object of class hglmm as produced by hglmm().

nsim

The number of data sets to simulate, from which to estimate parameters. From each data set a vector of parameters is estimated; the estimated covariance matrix is the empirical covariance matrix of these nsim vectors.

itmax

The maximum number of iterations to be used in attempting to achieve convergence when fitting models to the simulated data sets. Note that if convergence is not achieved, the simulated data set being used is discarded (i.e. it “doesn't count”) and a replacement data set is simulated.

verbose

Logical scalar. Should a progress report be printed out at each step of the fitting procedure? (Probably not, if the fitted procedure is to be undertaken 500 times.)

data.name

An identifying name to be included in the returned value. Defaults to object$data.name.

Details

Although this documentation refers to “generalized linear models”, currently only log linear Poisson models are provided for. More flexibility may be added at a future date.

Value

A list with components:

C.hat

The parametric bootstrap estimate of the covariance matrix of the parameter estimates.

nc.count

A count of the total number of times that the algorithm failed to converge during the bootstrapping procedure.

data.name

An identifying name, perhaps specifying the data set to which the model was fitted so as to produce object.

Author(s)

Rolf Turner r.turner@auckland.ac.nz

References

See the help for hglmm() for references.

See Also

fitted.hglmm() rhglmm() rhglmm.default() rhglmm.hglmm()

Examples

    ## Not run: 
        loc4 <- c("LngRf","BondiE","BondiOff","MlbrOff")
        SCC4 <- SydColCount[SydColCount$locn %in% loc4,] 
        SCC4$locn <- factor(SCC4$locn) # Get rid of unused levels.
        rownames(SCC4) <- 1:nrow(SCC4)
        fit <- hglmm(y~locn+depth,SCC4,c("locn","depth"),K=2,
                     contr="sum",verb=TRUE)
	cvr <- bcov(fit,nsim=500)
    
## End(Not run)

[Package hglmm version 0.0-14 Index]