Forecasting in Nonparametric Regression Models with Double Censoring

Resumen

In this work, we study the nonparametric estimation of the regression function using the least squares method in the presence of double censoring.

We construct an estimator by replacing unknown survival functions with self-consistent estimators in the spirit of Turnbull (1974).

We prove that this estimator is strongly consistent, converging almost surely to the optimal regression function.

Finally, we illustrate our theoretical findings with a simulation study under linear and nonlinear regression models.