A1156
Title: On least squares estimation and least absolute deviation estimation of trend models with time series errors
Authors: Qi Zheng - University of Louisville (United States)
Yunwei Cui - Towson University (United States) [presenting]
Abstract: The aim is to study the nonparametric estimation of trend functions with ARMA errors using a spline-based approach. Rather than employing the previous two-step procedure of estimating the trend and ARMA components separately, a unified one-step framework is proposed that jointly estimates the mean function and the ARMA parameters. The large-sample properties of both least squares estimation (LSE) and least absolute deviation estimation (LAD) are developed. To assess and compare their performances under various innovation distributions, an extensive simulation study is conducted. Finally, the practical utility of the proposed method is in using a real-world dataset.
