lfe (version 2.9-0)

btrap: Bootstrap standard errors for the group fixed effects

Description

Bootstrap standard errors for the group fixed effects which were swept out during an estimation with felm().

Usage

btrap(
  alpha,
  obj,
  N = 100,
  ef = NULL,
  eps = getOption("lfe.eps"),
  threads = getOption("lfe.threads"),
  robust = FALSE,
  cluster = NULL,
  lhs = NULL
)

Value

A data-frame of the same size as alpha is returned, with standard errors filled in.

Arguments

alpha

data frame returned from getfe()

obj

object of class "felm", usually, a result of a call to felm()

N

integer. The number of bootstrap iterations

ef

function. An estimable function such as in getfe(). The default is to use the one used on alpha

eps

double. Tolerance for centering, as in getfe

threads

integer. The number of threads to use

robust

logical. Should heteroskedastic standard errors be estimated?

cluster

logical or factor. Estimate clustered standard errors.

lhs

character vector. Specify which left hand side if obj has multiple lhs.

Details

The bootstrapping is done in parallel if threads > 1. btrap() is run automatically from getfe() if se=TRUE is specified. To save some overhead, the individual iterations are grouped together, the memory available for this grouping is fetched with getOption('lfe.bootmem'), which is initialized upon loading of lfe to options(lfe.bootmem=500) (MB).

If robust=TRUE, heteroskedastic robust standard errors are estimated. If robust=FALSE and cluster=TRUE, clustered standard errors with the cluster specified to felm() are estimated. If cluster is a factor, it is used for the cluster definition. cluster may also be a list of factors.

Examples

oldopts <- options("lfe.threads")
options(lfe.threads = 2)
## create covariates
x <- rnorm(3000)
x2 <- rnorm(length(x))

## create individual and firm
id <- factor(sample(700, length(x), replace = TRUE))
firm <- factor(sample(300, length(x), replace = TRUE))

## effects
id.eff <- rlnorm(nlevels(id))
firm.eff <- rexp(nlevels(firm))

## left hand side
y <- x + 0.25 * x2 + id.eff[id] + firm.eff[firm] + rnorm(length(x))

## estimate and print result
est <- felm(y ~ x + x2 | id + firm)
summary(est)
## extract the group effects
alpha <- getfe(est)
head(alpha)
## bootstrap standard errors
head(btrap(alpha, est))

## bootstrap some differences
ef <- function(v, addnames) {
  w <- c(v[2] - v[1], v[3] - v[2], v[3] - v[1])
  if (addnames) {
    names(w) <- c("id2-id1", "id3-id2", "id3-id1")
    attr(w, "extra") <- list(note = c("line1", "line2", "line3"))
  }
  w
}
# check that it's estimable
is.estimable(ef, est$fe)

head(btrap(alpha, est, ef = ef))
options(oldopts)