A non-Gaussian model for time series with pulses

Peter J. Diggle; Scott L. Zeger

doi:10.1080/01621459.1989.10478779

A non-Gaussian model for time series with pulses

Peter J. Diggle, Scott L. Zeger

Bloomberg School of Public Health

Research output: Contribution to journal › Article › peer-review

29 Scopus citations

Abstract

A non-Gaussian autoregressive-like model is presented for time series that exhibit occasional large increases in value, termed pulses, and exponential decay between pulses. The model differs from a first-order autoregressive process in its incorporation of feedback between the distribution of the current innovation and the history of the process. Likelihood-based methods of inference for the model are developed, and an application to endocrinological data is given.

Original language	English (US)
Pages (from-to)	354-359
Number of pages	6
Journal	Journal of the American Statistical Association
Volume	84
Issue number	406
DOIs	https://doi.org/10.1080/01621459.1989.10478779
State	Published - Jun 1989

Keywords

Autoregressive process
Feedback
Luteinizing hormone
Mixture distribution
Non-Gaussian time series

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1080/01621459.1989.10478779

Cite this

@article{e22731df81e4423492686aa4dd83d8b1,

title = "A non-Gaussian model for time series with pulses",

abstract = "A non-Gaussian autoregressive-like model is presented for time series that exhibit occasional large increases in value, termed pulses, and exponential decay between pulses. The model differs from a first-order autoregressive process in its incorporation of feedback between the distribution of the current innovation and the history of the process. Likelihood-based methods of inference for the model are developed, and an application to endocrinological data is given.",

keywords = "Autoregressive process, Feedback, Luteinizing hormone, Mixture distribution, Non-Gaussian time series",

author = "Diggle, {Peter J.} and Zeger, {Scott L.}",

note = "Funding Information: • Peter J. Diggle is Professor, Department of Mathematic~, Lanca~ter University, Lancaster LA1 4YL, England. Scott L.. Zeger. IS A:ssoclate Professor Department of Biostatistics, Johns Hopkins University, Baltimore, MD 20215. The authors thank M. Radford for providing his data, C. C. Law for computational assistance, and L. Ewing for helpful discussions on endocrinology. The associate editor and referee are also thanked for their insightful comments. Support for this rese~rch. from the Commonwealth Scientific and Industrial Research Organisation ~f Australia and National Institutes of Health Grant R29-AI-25529-02 IS gratefully acknowledged.",

year = "1989",

month = jun,

doi = "10.1080/01621459.1989.10478779",

language = "English (US)",

volume = "84",

pages = "354--359",

journal = "Journal of the American Statistical Association",

issn = "0162-1459",

publisher = "Taylor and Francis Ltd.",

number = "406",

}

TY - JOUR

T1 - A non-Gaussian model for time series with pulses

AU - Diggle, Peter J.

AU - Zeger, Scott L.

N1 - Funding Information: • Peter J. Diggle is Professor, Department of Mathematic~, Lanca~ter University, Lancaster LA1 4YL, England. Scott L.. Zeger. IS A:ssoclate Professor Department of Biostatistics, Johns Hopkins University, Baltimore, MD 20215. The authors thank M. Radford for providing his data, C. C. Law for computational assistance, and L. Ewing for helpful discussions on endocrinology. The associate editor and referee are also thanked for their insightful comments. Support for this rese~rch. from the Commonwealth Scientific and Industrial Research Organisation ~f Australia and National Institutes of Health Grant R29-AI-25529-02 IS gratefully acknowledged.

PY - 1989/6

Y1 - 1989/6

N2 - A non-Gaussian autoregressive-like model is presented for time series that exhibit occasional large increases in value, termed pulses, and exponential decay between pulses. The model differs from a first-order autoregressive process in its incorporation of feedback between the distribution of the current innovation and the history of the process. Likelihood-based methods of inference for the model are developed, and an application to endocrinological data is given.

AB - A non-Gaussian autoregressive-like model is presented for time series that exhibit occasional large increases in value, termed pulses, and exponential decay between pulses. The model differs from a first-order autoregressive process in its incorporation of feedback between the distribution of the current innovation and the history of the process. Likelihood-based methods of inference for the model are developed, and an application to endocrinological data is given.

KW - Autoregressive process

KW - Feedback

KW - Luteinizing hormone

KW - Mixture distribution

KW - Non-Gaussian time series

UR - http://www.scopus.com/inward/record.url?scp=0001291561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001291561&partnerID=8YFLogxK

U2 - 10.1080/01621459.1989.10478779

DO - 10.1080/01621459.1989.10478779

M3 - Article

AN - SCOPUS:0001291561

SN - 0162-1459

VL - 84

SP - 354

EP - 359

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

IS - 406

ER -

A non-Gaussian model for time series with pulses

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this