clusterSEs (version 2.2)

cluster.bs.plm: Pairs Cluster Bootstrapped p-Values For PLM

Description

This software estimates p-values using pairs cluster bootstrapped t-statistics for fixed effects panel linear models (Cameron, Gelbach, and Miller 2008). The data set is repeatedly re-sampled by cluster, a model is estimated, and inference is based on the sampling distribution of the pivotal (t) statistic.

Usage

cluster.bs.plm(mod, dat, cluster = "group", ci.level = 0.95,
  boot.reps = 1000, cluster.se = TRUE, report = TRUE, prog.bar = TRUE)

Arguments

mod

A "within" model estimated using plm.

dat

The data set used to estimate mod.

cluster

Clustering dimension ("group", the default, or "time").

ci.level

What confidence level should CIs reflect?

boot.reps

The number of bootstrap samples to draw.

cluster.se

Use clustered standard errors (= TRUE) or ordinary SEs (= FALSE) for bootstrap replicates.

report

Should a table of results be printed to the console?

prog.bar

Show a progress bar of the bootstrap (= TRUE) or not (= FALSE).

Value

• A list with the elements
• p.values
  A matrix of the estimated p-values.
• ci
  A matrix of confidence intervals.

References

Cameron, A. Colin, Jonah B. Gelbach, and Douglas L. Miller. 2008. "Bootstrap-Based Improvements for Inference with Clustered Errors." The Review of Economics and Statistics 90(3): 414-427.

Examples

# predict employment levels, cluster on group
require(plm)
data(EmplUK)

emp.1 <- plm(emp ~ wage + log(capital+1), data = EmplUK, 
             model = "within", index=c("firm", "year"))
cluster.bs.plm(mod=emp.1, dat=EmplUK, cluster="group", ci.level = 0.95, 
          boot.reps = 1000, cluster.se = TRUE, report = TRUE, 
          prog.bar = TRUE)

# cluster on time

cluster.bs.plm(mod=emp.1, dat=EmplUK, cluster="time", ci.level = 0.95, 
            boot.reps = 1000, cluster.se = TRUE, report = TRUE, 
            prog.bar = TRUE)