Discussion of “High-dimensional autocovariance matrices and optimal linear prediction”

Research output: Contribution to journalArticleResearchpeer-review

Abstract

I propose new ACF and PACF plots based on the autocovari- ance estimators of McMurry and Politis. I also show that the forecasting methods they propose perform poorly compared to some relatively simple autoregression algorithms already available.
Original languageEnglish
Pages (from-to)792-796
Number of pages5
JournalElectronic Journal of Statistics
Volume9
Issue number1
DOIs
Publication statusPublished - 2015

Cite this

@article{b856fa4f2ea8423cb8ca8a35042b4cb5,
title = "Discussion of “High-dimensional autocovariance matrices and optimal linear prediction”",
abstract = "I propose new ACF and PACF plots based on the autocovari- ance estimators of McMurry and Politis. I also show that the forecasting methods they propose perform poorly compared to some relatively simple autoregression algorithms already available.",
author = "Hyndman, {Robin John}",
year = "2015",
doi = "10.1214/14-EJS953",
language = "English",
volume = "9",
pages = "792--796",
journal = "Electronic Journal of Statistics",
issn = "1935-7524",
publisher = "Institute of Mathematical Statistics",
number = "1",

}

Discussion of “High-dimensional autocovariance matrices and optimal linear prediction”. / Hyndman, Robin John.

In: Electronic Journal of Statistics, Vol. 9, No. 1, 2015, p. 792-796.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Discussion of “High-dimensional autocovariance matrices and optimal linear prediction”

AU - Hyndman, Robin John

PY - 2015

Y1 - 2015

N2 - I propose new ACF and PACF plots based on the autocovari- ance estimators of McMurry and Politis. I also show that the forecasting methods they propose perform poorly compared to some relatively simple autoregression algorithms already available.

AB - I propose new ACF and PACF plots based on the autocovari- ance estimators of McMurry and Politis. I also show that the forecasting methods they propose perform poorly compared to some relatively simple autoregression algorithms already available.

U2 - 10.1214/14-EJS953

DO - 10.1214/14-EJS953

M3 - Article

VL - 9

SP - 792

EP - 796

JO - Electronic Journal of Statistics

JF - Electronic Journal of Statistics

SN - 1935-7524

IS - 1

ER -