SBdecomp (version 1.2)

pred.smooth: Nonparametric conditional mean

Description

Calculates expected outcome given a single predictor (conditional mean) using a nonparametric kernel estimator

Usage

pred.smooth(zz, zi.one, bw = NULL, y1, weight = NULL)

Arguments

zz

predictor; conditional mean function is applied to this predictor to produce predicted outcome

zi.one

predictor; used to estimate conditional mean function (often same as zz)

bw

bandwidth

y1

outcome

weight

weight

Value

expected outcome given predictor

References

Parast, L and Griffin, BA (2020). "Quantifying the Bias due to Observed Individual Confounders in Causal Treatment Effect Estimates." Statistics in Medicine, In press.