oos_partial: Estimate matrix of bivariate out-of-sample variance explained

In akcochrane/ACmisc: Aaron Cochrane's miscellaneous functions

Estimate matrix of bivariate out-of-sample variance explained

Description

Uses tef_rlm_boot() to get out-of-sample variance explained as well as bootstrapped p values. The latter is two times the proportion of the bootstrapped distribution is on the opposite side of zero than the median coefficient.

Usage

oos_partial(xvars, partialvars = c(), data, nBoots = 200)

Arguments

xvars	Vector of variable names, with which to estimate bivariate relations.
partialvars	If a vector of variable names are provided, then the lower triangle of the output is the out-of-sample variance explained when controlling for the variables in partialvars.
data	Data to use
nBoots	Number of resamples

Details

Returns a list with two matrices: The first reports the bivariate out-of-sample variance explained (upper triangle) and partial out-of-sample variance explained (lower triangle). The second reports the pseudo-p-values, as described above.

akcochrane/ACmisc documentation built on Nov. 24, 2024, 11:22 a.m.