predict.CEAforests: Predict with a CEA forest.

In bonander/CEAforests: Tree-based heterogenous cost-effectiveness and policy learning with individual-level data (beta)

Description

predict.CEAforests Gets estimates of conditional incremental outcomes and costs given x using a cea_forest object (a wrapper for grf::predict.causal_forest).

Usage

## S3 method for class 'CEAforests'
predict(object, ...)

Arguments

object	The trained CEA forest.
...	Other options to be passed to grf::predict.causal_forest() or grf::predict.instrumental_forest(). See grf documentation for additional information.

Value

A matrix of predictions of conditional average treatment effects for the outcome and costs, along with variance estimates. Also returns debiased errors (estimates of the error of a forest with infinite size) and excess error due to Monte Carlo variability (estimated via jackknife). The latter provides an estimates of how unstable the estimates are if we grow forests of the same size on the same dataset. Increase the number of trees until the excess error becomes negligible. See grf::predict.causal_forest documentation for further details.

Examples

## Not run: 
To be added...

## End(Not run)

bonander/CEAforests documentation built on April 1, 2021, 10:57 a.m.