ecap (version 0.1.2)

predict.ecap: Implementing the ECAP procedure

Description

Takes in an ECAP object and a new set of probability estimates that the user wishes to adjust. The model uses the calibration from the ecap object to ECAP adjust the new probability estimates given to the function predict.

Usage

# S3 method for ecap
predict(object, new_unadjusted, ...)

Arguments

object

An object of class ecap.

new_unadjusted

A numerical vector of unadjusted probabilities that you want to ECAP adjust.

...

Additional arguments

Value

A vector of ECAP adjusted probability estimates.

References

http://faculty.marshall.usc.edu/gareth-james/Research/Probs.pdf

Examples

# NOT RUN {
set.seed(1)
p_obs <- runif(1000, 0, 1)
win_var <- rbinom(length(p_obs), 1, p_obs)
ecap_fit <- ecap(unadjusted_prob = p_obs, win_var = win_var, win_id = 1, bias_indicator = FALSE)

p_new <- runif(1000, 0, 1)
ecap_new <- predict(object=ecap_fit, new_unadjusted=p_new)
# }