error_prop: Estimate Errors of Regression Coefficients

In cape: Combined Analysis of Pleiotropy and Epistasis for Diversity Outbred Mice

Estimate Errors of Regression Coefficients

Description

This function uses error propagation formulas for quantities computed from regression coefficients to estimate the error for all regression coefficients.

Usage

error_prop(
  data_obj,
  pairscan_obj,
  perm = FALSE,
  verbose = FALSE,
  run_parallel = FALSE,
  n_cores = 4,
  just_m = FALSE
)

Arguments

data_obj	a Cape object
pairscan_obj	a pairscan object from pairscan
perm	A logical value to indicate whether error propagation should be performed on the test statistics (FALSE) or the permuted test statistics (TRUE).
verbose	A logical value to indicate whether the progress of the function should be printed to the screen.
run_parallel	boolean, default = FALSE
n_cores	The number of cores to use if run_parallel is TRUE, default = 4
just_m	If TRUE only the m12 and m21 values are calculated. If FALSE, the default, the standard deviations are also calculated.

Value

This function returns the data object with a new list element: var_to_var_influences if perm is set to FALSE and var_to_var_influences_perm if perm is set to TRUE. These tables include the errors calculated for the marker1 to marker2 (m21) influences as well as the marker2 to marker1 (m12) influences. These results are used by calc_p to calculate empirical p values.

Examples

## Not run: 
#run error propagateion on test statistics and 
#permuted test statistics
data_obj <- error_prop(data_obj, pairscan_obj, perm = TRUE)
data_obj <- error_prop(data_obj, pairscan_obj, perm = FALSE)

## End(Not run)

cape documentation built on May 20, 2022, 1:06 a.m.