NRejections (version 1.2.0)

resample_resid: Resample residuals for OLS

Description

Implements the residual resampling OLS algorithm described in Mathur & VanderWeele (in preparation). Specifically, the design matrix is fixed while the resampled outcomes are set equal to the original fitted values plus a vector of residuals sampled with replacement.

Usage

resample_resid(
  d,
  X,
  C = NA,
  Ys,
  alpha,
  resid,
  bhat.orig,
  B = 2000,
  cores = NULL
)

Arguments

d

Dataframe

X

Single quoted name of covariate of interest

C

Vector of quoted covariate names

Ys

Vector of quoted outcome names

alpha

Alpha level for individual tests

resid

Residuals from original sample (W X B matrix)

bhat.orig

Estimated coefficients for covariate of interest in original sample (W-vector)

B

Number of resamples to generate

cores

Number of cores available for parallelization

Value

Returns a list containing the number of rejections in each resample, a matrix of p-values in the resamples, and a matrix of t-statistics in the resamples.

References

Mathur, M.B., & VanderWeele, T.J. (in preparation). New metrics for multiple testing with correlated outcomes.

Examples

# NOT RUN {
samp.res = dataset_result( X = "complaints",
                C = c("privileges", "learning"),
                Ys = c("rating", "raises"),
                d = attitude,
                center.stats = FALSE,
                bhat.orig = NA,  # bhat.orig is a single value now for just the correct Y
                alpha = 0.05 )

resamps = resample_resid(  X = "complaints",
                  C = c("privileges", "learning"),
                  Ys = c("rating", "raises"),
                  d = attitude,
                  alpha = 0.05,
                  resid = samp.res$resid,
                  bhat.orig = samp.res$b,
                  B=20,
                  cores = 2)
# }