Conditional Density Kernel Estimation Under Random Censorship for Functional Weak Dependence Data

The primary objective of this research is to investigate the asymptotic properties of the conditional density nonparametric estimator. The main areas of focus are the estimator’s consistency (with rates), including those involving censored data and quasi-associated dependent variables, as well as its performance when the covariate is functional in nature. For this model, we establish the almost complete pointwise convergence of the conditional density estimate. The findings from this research contribute to the theoretical foundations of nonparametric density estimation, with direct implications for data analysis and decision-making in various fields, such as biomedical research, finance, and social sciences. To empirically examine the practical implications of the established asymptotic properties, we conducted a series of simulation experiments. These numerical studies allow us to investigate the finite-sample performance of the conditional density nonparametric estimator and validate the theoretical findings.