Online Program Home
My Program

Abstract Details

Activity Number: 345 - Theory and Methods for Multivariate Analysis
Type: Contributed
Date/Time: Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
Sponsor: IMS
Abstract #330544 Presentation
Title: A Semiparametric Approach for Modeling Multivariate Nonlinear Time Series
Author(s): Seyed Yaser Samadi*
Companies: Southern Illinois University, Carbondale
Keywords: Multivariate nonlinear time series; Semiparametric estimation; Kernel density estimation; L2-fitting criterion;
Abstract:

Multivariate time series data are ubiquitous arising from myriad applications. In recent years, a combination of parametric and nonparametric estimates have been used to make a more efficient model in different fields of sciences. We propose a proper combination of parametric and nonparametric methods to estimate the multivariate nonlinear autoregressive function of multivariate nonlinear time series. First, we estimate the link function by considering the multivariate Taylor series expansion of the link function. Then, we adjusted our initial estimation by a nonparametric diagonal matrix. The matrix of adjustment factors are defined as the minimizers of the L2-fitting criterion In other word, we use the local fitting approach to estimate the adjustment factor by using a kernel-smoothed function, which locally approximate the true density function at each local point. The asymptotic consistency properties of the proposed estimators are established. Simulation studies and a real data analysis are conducted to evaluate the performance of the proposed semiparametric approach.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program