Online Program Home
My Program

Abstract Details

Activity Number: 64 - BandE Student Paper Awards
Type: Topic Contributed
Date/Time: Sunday, July 29, 2018 : 4:00 PM to 5:50 PM
Sponsor: Business and Economic Statistics Section
Abstract #327009
Title: Partial Distance Correlation Screening for High-Dimensional Time Series
Author(s): Kashif Yousuf* and Yang Feng
Companies: Columbia University and Columbia University
Keywords: Time Series; High dimensional Statistics; Feature Screening; Distance Correlation; Sure Independence Screening; Functional Dependence Measure

High dimensional time series datasets are becoming increasingly common in various fields such as economics, finance, meteorology, and neuroscience. Given this ubiquity of time series data, it is surprising that very few works on variable screening are applicable to time dependent data, and even fewer methods are developed which utilize the unique aspects of time series data. This paper introduces several model free screening procedures developed specifically to deal with dependent and/or heavy tailed response and covariate time series. These methods are based on the partial distance correlation, and are developed both for univariate response models, such as nonlinear autoregressive models with exogenous predictors, and multivariate response models such as linear or nonlinear VAR models. Sure screening properties are proved for our procedures, which depend on the moment conditions, and the strength of dependence in the response and covariate processes, amongst other factors. Finite sample performance of these procedures is shown through extensive simulation studies, and we include an application to macroeconomic forecasting.

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

Back to the full JSM 2018 program