The conditional variance function in a heteroscedastic, nonparametric regression model is estimated by linear smoothing of squared residuals. Attention is focused on local polynomial smoothers. Both ...

We suggest a one-dimensional jump-detection algorithm based on local polynomial fitting for jumps in regression functions (zero-order jumps) or jumps in derivatives (first-order or higher-order jumps) ...

Abstract. We present a nonparametric approach based on local polynomial regression for ensemble forecast of time series. The state space is first reconstructed by embedding the univariate time series ...