The npcure R package provides nonparametric methods for estimating cure probability and latency functions in mixture cure models. It allows for the estimation of these functions without assuming a specific parametric form, making it suitable for survival data with a cure fraction. The package includes nonparametric estimators for the cure probability and latency function, along with bootstrap bandwidth selectors for these estimators. It also implements a nonparametric test for the significance of covariates in the cure probability. The cure probability is estimated using a nonparametric kernel-type estimator, while the latency function is estimated using a nonparametric estimator based on the conditional survival function. Both estimators are consistent under certain conditions, and their performance is evaluated using bootstrap methods for bandwidth selection. The package also includes a nonparametric test for the significance of covariates in the cure probability, which can be applied to continuous, discrete, or qualitative covariates. The npcure package is designed for one-dimensional continuous covariates and uses the Epanechnikov kernel for smoothing. It provides functions for estimating the cure rate, latency function, and conditional survival function, as well as for performing significance tests for covariates. The package also includes functions for bandwidth selection and for computing confidence intervals for the estimated parameters. The package is implemented in R and is available from CRAN.The npcure R package provides nonparametric methods for estimating cure probability and latency functions in mixture cure models. It allows for the estimation of these functions without assuming a specific parametric form, making it suitable for survival data with a cure fraction. The package includes nonparametric estimators for the cure probability and latency function, along with bootstrap bandwidth selectors for these estimators. It also implements a nonparametric test for the significance of covariates in the cure probability. The cure probability is estimated using a nonparametric kernel-type estimator, while the latency function is estimated using a nonparametric estimator based on the conditional survival function. Both estimators are consistent under certain conditions, and their performance is evaluated using bootstrap methods for bandwidth selection. The package also includes a nonparametric test for the significance of covariates in the cure probability, which can be applied to continuous, discrete, or qualitative covariates. The npcure package is designed for one-dimensional continuous covariates and uses the Epanechnikov kernel for smoothing. It provides functions for estimating the cure rate, latency function, and conditional survival function, as well as for performing significance tests for covariates. The package also includes functions for bandwidth selection and for computing confidence intervals for the estimated parameters. The package is implemented in R and is available from CRAN.