AUC {hdcuremodels}

AUC for cure prediction using mean score imputation

Description

This function calculates the AUC for cure prediction using the mean score imputation (MSI) method proposed by Asano et al.

Usage

AUC(object, newdata, cure_cutoff = 5, model.select = "AIC")

Arguments

object

a mixturecure object resulting from curegmifs, cureem, cv_curegmifs, cv_cureem.

newdata

an optional data.frame that minimally includes the incidence and/or latency variables to use for predicting the response. If omitted, the training data are used.

cure_cutoff

cutoff value for cure, used to produce a proxy for the unobserved cure status; default is 5.

model.select

for models fit using curegmifs or cureem any step along the solution path can be selected. The default is model.select = "AIC" which calculates the predicted values using the coefficients from the model having the lowest AIC. Other options are model.select = "mAIC" for the modified AIC, model.select = "cAIC" for the corrected AIC, model.select = "BIC", model.select = "mBIC" for the modified BIC, model.select = "EBIC" for the extended BIC, model.select = "logLik" for the step that maximizes the log-likelihood, or any numeric value from the solution path. This option has no effect for objects fit using cv_curegmifs or cv_cureem.

Value

Returns the AUC value for cure prediction using the mean score imputation (MSI) method.

References

Asano, J., Hirakawa, H., Hamada, C. (2014) Assessing the prediction accuracy of cure in the Cox proportional hazards cure model: an application to breast cancer data. Pharmaceutical Statistics, 13:357–363.

See Also

concordance_mcm

Examples

library(survival)
set.seed(1234)
temp <- generate_cure_data(N = 100, J = 10, nTrue = 10, A = 1.8)
training <- temp$Training
testing <- temp$Testing
fit <- curegmifs(Surv(Time, Censor) ~ .,
                  data = training, x.latency = training,
                  model = "weibull", thresh = 1e-4, maxit = 2000,
                  epsilon = 0.01, verbose = FALSE)
AUC(fit)
AUC(fit, newdata = testing)