Forecasting with Functional and Twice Censored Data

In this study, we propose a new kernel functional regression estimator when the random response variable is subject to twice censoring. Censoring is employed to handle cases where complete response data is unavailable, allowing for more robust and reliable statistical analysis. Our proposed estimator is specifically designed to provide accurate forecasts even in the presence of such incomplete data. Then, we investigate its mean square convergence, with rate. To reinforce the obtained results, we conduct numerical results to highlight the performance and the accuracy of our proposed estimator.