trim_ps {beastt}R Documentation

Trim a prop_scr object

Description

Trim a prop_scr object

Usage

trim_ps(x, low = NULL, high = NULL, quantile = FALSE)

Arguments

x

A prop_scr object

low

Low cut-off such that all participants with propensity scores less than this value (or quantile if quantile = TRUE) are removed. If left NULL no lower bound will be used

high

High cut-off such that all participants with propensity scores greater than this value (or quantile if quantile = TRUE) are removed. If left NULL no upper bound will be used

quantile

True/False value to determine if the cut-off values are based directly on the propensity scores (false) or their quantiles (true). By default this is false.

Details

This function uses R's default method of quantile calculation (type 7)

Value

a prop_scr object with a trimmed propensity score distribution

Examples

library(dplyr)
ps_obj <- calc_prop_scr(internal_df = filter(int_binary_df, trt == 0),
                       external_df = ex_binary_df,
                       id_col = subjid,
                       model = ~ cov1 + cov2 + cov3 + cov4)
trim_ps(ps_obj, low = 0.3, high = 0.7)


[Package beastt version 0.0.3 Index]