High-Dimensional Knockoffs Inference for Time Series Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chi, Chien-Ming, Fan, Yingying, Ing, Ching-Kang, Lv, Jinchi
Format: Preprint
Published: 2021
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917939681165312
author Chi, Chien-Ming
Fan, Yingying
Ing, Ching-Kang
Lv, Jinchi
author_facet Chi, Chien-Ming
Fan, Yingying
Ing, Ching-Kang
Lv, Jinchi
contents We make some initial attempt to establish the theoretical and methodological foundation for the model-X knockoffs inference for time series data. We suggest the method of time series knockoffs inference (TSKI) by exploiting the ideas of subsampling and e-values to address the difficulty caused by the serial dependence. We also generalize the robust knockoffs inference in Barber, Candès, and Samworth to the time series setting to relax the assumption of known covariate distribution required by model-X knockoffs, since such an assumption is overly stringent for time series data. We establish sufficient conditions under which TSKI achieves the asymptotic false discovery rate (FDR) control. Our technical analysis reveals the effects of serial dependence and unknown covariate distribution on the FDR control. We conduct a power analysis of TSKI using the Lasso coefficient difference knockoff statistic under the generalized linear time series models. The finite-sample performance of TSKI is illustrated with several simulation examples and an economic inflation study.
format Preprint
id arxiv_https___arxiv_org_abs_2112_09851
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle High-Dimensional Knockoffs Inference for Time Series Data
Chi, Chien-Ming
Fan, Yingying
Ing, Ching-Kang
Lv, Jinchi
Methodology
Statistics Theory
62P20, Stats.ML
A.0
We make some initial attempt to establish the theoretical and methodological foundation for the model-X knockoffs inference for time series data. We suggest the method of time series knockoffs inference (TSKI) by exploiting the ideas of subsampling and e-values to address the difficulty caused by the serial dependence. We also generalize the robust knockoffs inference in Barber, Candès, and Samworth to the time series setting to relax the assumption of known covariate distribution required by model-X knockoffs, since such an assumption is overly stringent for time series data. We establish sufficient conditions under which TSKI achieves the asymptotic false discovery rate (FDR) control. Our technical analysis reveals the effects of serial dependence and unknown covariate distribution on the FDR control. We conduct a power analysis of TSKI using the Lasso coefficient difference knockoff statistic under the generalized linear time series models. The finite-sample performance of TSKI is illustrated with several simulation examples and an economic inflation study.
title High-Dimensional Knockoffs Inference for Time Series Data
topic Methodology
Statistics Theory
62P20, Stats.ML
A.0
url https://arxiv.org/abs/2112.09851