predict.CausalStump: Predict the response surfaces using a Gaussian process or...

In mazphilip/CausalStumps: CausalStump: A structural Gaussian process implementation for Causal Inference

Description

Predict the response surfaces using a Gaussian process or Student-t process fit.

Usage

## S3 method for class 'CausalStump'
predict(cs_object, X, z, pscore, nsampling)

Arguments

cs_object	A object from the CausalStump function
X	A data.frame with new data. If not presented, using the training data.
z	A vector or scalar with new treatment data. If it is a scalar, it predicts using the same value for all observations. If missing, it uses the training data.
pscore	A vector with the propensity score. Throws an error if use is different to the CausalStump fit.
nsampling	(optional) A number that overwrites the number of samples used for the TP-kernel sampling.

Value

Returns the MAP and the 95 percent credible interval of the fitted process as a list.

mazphilip/CausalStumps documentation built on May 12, 2019, 7:22 p.m.