plmm.bs: Bootstrap Inference

In plmm: Partially Linear Mixed Effects Model

Description

The fixed regression coefficients and the random effects variance are repeatedly estimated using the wild bootstrap to estimate the sampling distribution of the estimators.

Usage

plmm.bs(object, B, data, h0, ...)

Arguments

object	a model estimated using a model fitting function plmm or wplmm.
B	the number of bootstrap replications.
data	an optional data frame containing the variables in the model. If relevant variables are not found in data, the variables are taken from the environment from which plmm.bs was called.
h0	a set of bandwidths for nonparametric kernel regression to reduce the model. h0 is optional; if omitted, select.h0 is called automatically to compute bandwidths according to the h0.call value of the object.
...	other arguments.

Details

plmm.bs computes bootstrap estimates of the fixed regression coefficients and the random effects variance using the wild bootstrap resampling technique. The user can modify bandwidths in a list object created by select.h0 and pass the object to plmm.bs. In the wild bootstrap procedure, only the bandwidth for the kernel estimation of the response is recalculated while those for the fixed parametric component remain the same as given in h0. plmm.bs returns an object of bs.plmm class for which the summary method is available.

Value

A list of a matrix of B rows. Its columns contain B bootstrap estimates of the fixed regression coefficients and random effects variance.

See Also

summary.bs.plmm, select.h0.

Examples

data(plmm.data)
h0 <- select.h0(y1~x1+x2+x3|t1, data=plmm.data)
model <- plmm(y1~x1+x2+x3|t1, h0=h0, random=cluster, data=plmm.data)
model2 <- wplmm(model, heteroX=x3, data=plmm.data)

## Not run: 
bs <- plmm.bs(model, B=500, data=plmm.data, h0=h0)
bs2 <- plmm.bs(model2, B=500, data=plmm.data, h0=h0) 
## End(Not run)

plmm documentation built on May 2, 2019, 7:29 a.m.