Inference for modulated stationary processes

Zhibiao Zhao, Xiaoye Li

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

We study statistical inferences for a class of modulated stationary processes with time-dependent variances.Due to non-stationarity and the large number of unknown parameters, existing methods for stationary, or locally stationary, time series are not applicable. Based on a self-normalization technique, we address several inference problems, including a self-normalized central limit theorem, a self-normalized cumulative sum test for the change-point problem, a long-run variance estimation through blockwise self-normalization, and a self-normalization-based wild bootstrap. Monte Carlo simulation studies show that the proposed self-normalization-based methods outperform stationarity-based alternatives.We demonstrate the proposed methodology using two real data sets: annual mean precipitation rates in Seoul from 1771-2000, and quarterly U.S. Gross National Product growth rates from 1947-2002.

Original languageEnglish (US)
Pages (from-to)205-227
Number of pages23
JournalBernoulli
Volume19
Issue number1
DOIs
StatePublished - Feb 1 2013

Fingerprint

Self-normalization
Stationary Process
Self-normalized Sums
Wild Bootstrap
Change-point Problem
Stationary Time Series
Cumulative Sum
Nonstationarity
Variance Estimation
Stationarity
Long-run
Statistical Inference
Gross
Central limit theorem
Unknown Parameters
Annual
Monte Carlo Simulation
Simulation Study
Methodology
Alternatives

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

Zhao, Zhibiao ; Li, Xiaoye. / Inference for modulated stationary processes. In: Bernoulli. 2013 ; Vol. 19, No. 1. pp. 205-227.
@article{7e966fbb33bd4bdb8e7b1b3ce2609548,
title = "Inference for modulated stationary processes",
abstract = "We study statistical inferences for a class of modulated stationary processes with time-dependent variances.Due to non-stationarity and the large number of unknown parameters, existing methods for stationary, or locally stationary, time series are not applicable. Based on a self-normalization technique, we address several inference problems, including a self-normalized central limit theorem, a self-normalized cumulative sum test for the change-point problem, a long-run variance estimation through blockwise self-normalization, and a self-normalization-based wild bootstrap. Monte Carlo simulation studies show that the proposed self-normalization-based methods outperform stationarity-based alternatives.We demonstrate the proposed methodology using two real data sets: annual mean precipitation rates in Seoul from 1771-2000, and quarterly U.S. Gross National Product growth rates from 1947-2002.",
author = "Zhibiao Zhao and Xiaoye Li",
year = "2013",
month = "2",
day = "1",
doi = "10.3150/11-BEJ399",
language = "English (US)",
volume = "19",
pages = "205--227",
journal = "Bernoulli",
issn = "1350-7265",
publisher = "International Statistical Institute",
number = "1",

}

Inference for modulated stationary processes. / Zhao, Zhibiao; Li, Xiaoye.

In: Bernoulli, Vol. 19, No. 1, 01.02.2013, p. 205-227.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Inference for modulated stationary processes

AU - Zhao, Zhibiao

AU - Li, Xiaoye

PY - 2013/2/1

Y1 - 2013/2/1

N2 - We study statistical inferences for a class of modulated stationary processes with time-dependent variances.Due to non-stationarity and the large number of unknown parameters, existing methods for stationary, or locally stationary, time series are not applicable. Based on a self-normalization technique, we address several inference problems, including a self-normalized central limit theorem, a self-normalized cumulative sum test for the change-point problem, a long-run variance estimation through blockwise self-normalization, and a self-normalization-based wild bootstrap. Monte Carlo simulation studies show that the proposed self-normalization-based methods outperform stationarity-based alternatives.We demonstrate the proposed methodology using two real data sets: annual mean precipitation rates in Seoul from 1771-2000, and quarterly U.S. Gross National Product growth rates from 1947-2002.

AB - We study statistical inferences for a class of modulated stationary processes with time-dependent variances.Due to non-stationarity and the large number of unknown parameters, existing methods for stationary, or locally stationary, time series are not applicable. Based on a self-normalization technique, we address several inference problems, including a self-normalized central limit theorem, a self-normalized cumulative sum test for the change-point problem, a long-run variance estimation through blockwise self-normalization, and a self-normalization-based wild bootstrap. Monte Carlo simulation studies show that the proposed self-normalization-based methods outperform stationarity-based alternatives.We demonstrate the proposed methodology using two real data sets: annual mean precipitation rates in Seoul from 1771-2000, and quarterly U.S. Gross National Product growth rates from 1947-2002.

UR - http://www.scopus.com/inward/record.url?scp=84883856432&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883856432&partnerID=8YFLogxK

U2 - 10.3150/11-BEJ399

DO - 10.3150/11-BEJ399

M3 - Article

VL - 19

SP - 205

EP - 227

JO - Bernoulli

JF - Bernoulli

SN - 1350-7265

IS - 1

ER -