performance_measure: A performance mesure function

In syhyunpark/csim: Constrained single index models (CSIM)

Description

performance_measure assesses the performance of a given treatment selection rule, in terms of the "Value" of the treatment selection rule and the proportion of correctly made decisions (PCD).

Usage

performance_measure(pred.test, y, Tr, value.opt = NULL, optTr = NULL)

Arguments

pred.test	a n x K matrix of predicted values given from any model; each column represents each treatment (trt 1, trt 2,..., trt K)
y	a n x 1 vector of treatment outcomes from a testing set
Tr	a n x 1 vector of treatment indicators from a testing set
value.opt	the optimal value of treatment selection rules for the testing set (only available for a simulated dataset).
optTr	a vector of treatments indicating the optimal treatment selections for the testing set (only available for a simulated dataset).

Value

a list of information contating the estimatd Value and the proportion of correct decisions.

See Also

dataGenerationFn

syhyunpark/csim documentation built on May 31, 2019, 4:56 a.m.