Independence, successive and conditional likelihood for time series of counts

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

Independence, successive and conditional likelihood for time series of counts. / Sørensen, Helle.

I: Journal of Statistical Planning and Inference, Bind 200, 2019, s. 20-31.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Sørensen, H 2019, 'Independence, successive and conditional likelihood for time series of counts', Journal of Statistical Planning and Inference, bind 200, s. 20-31. https://doi.org/10.1016/j.jspi.2018.09.002

APA

Sørensen, H. (2019). Independence, successive and conditional likelihood for time series of counts. Journal of Statistical Planning and Inference, 200, 20-31. https://doi.org/10.1016/j.jspi.2018.09.002

Vancouver

Sørensen H. Independence, successive and conditional likelihood for time series of counts. Journal of Statistical Planning and Inference. 2019;200:20-31. https://doi.org/10.1016/j.jspi.2018.09.002

Author

Sørensen, Helle. / Independence, successive and conditional likelihood for time series of counts. I: Journal of Statistical Planning and Inference. 2019 ; Bind 200. s. 20-31.

Bibtex

@article{5763fe73b7484c31a5eb34abf9d39d60,
title = "Independence, successive and conditional likelihood for time series of counts",
abstract = "Serial correlation and overdispersion must be handled properly in analyses of time series of counts, and parameter-driven models combine an underlying latent process with a conditional log-linear Poisson model (given the latent process) for that purpose. Regression coefficients have direct interpretations, but likelihood inference is not straight-forward. We consider a two-step procedure for estimation: First regression parameters are estimated from the marginal distribution; second parameters concerning the latent process are estimated with composite likelihood methods, based on low-order simultaneous or conditional distributions. Confidence intervals are computed by bootstrap. Properties of estimators are examined and compared to other methods in three simulation studies, and the methods are applied to two datasets from the literature concerning hospital admission related to asthma and traffic deaths.",
keywords = "Bootstrap, Composite likelihood, Generalized linear mixed model, Overdispersion, Serial correlation",
author = "Helle S{\o}rensen",
year = "2019",
doi = "10.1016/j.jspi.2018.09.002",
language = "English",
volume = "200",
pages = "20--31",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier BV * North-Holland",

}

RIS

TY - JOUR

T1 - Independence, successive and conditional likelihood for time series of counts

AU - Sørensen, Helle

PY - 2019

Y1 - 2019

N2 - Serial correlation and overdispersion must be handled properly in analyses of time series of counts, and parameter-driven models combine an underlying latent process with a conditional log-linear Poisson model (given the latent process) for that purpose. Regression coefficients have direct interpretations, but likelihood inference is not straight-forward. We consider a two-step procedure for estimation: First regression parameters are estimated from the marginal distribution; second parameters concerning the latent process are estimated with composite likelihood methods, based on low-order simultaneous or conditional distributions. Confidence intervals are computed by bootstrap. Properties of estimators are examined and compared to other methods in three simulation studies, and the methods are applied to two datasets from the literature concerning hospital admission related to asthma and traffic deaths.

AB - Serial correlation and overdispersion must be handled properly in analyses of time series of counts, and parameter-driven models combine an underlying latent process with a conditional log-linear Poisson model (given the latent process) for that purpose. Regression coefficients have direct interpretations, but likelihood inference is not straight-forward. We consider a two-step procedure for estimation: First regression parameters are estimated from the marginal distribution; second parameters concerning the latent process are estimated with composite likelihood methods, based on low-order simultaneous or conditional distributions. Confidence intervals are computed by bootstrap. Properties of estimators are examined and compared to other methods in three simulation studies, and the methods are applied to two datasets from the literature concerning hospital admission related to asthma and traffic deaths.

KW - Bootstrap

KW - Composite likelihood

KW - Generalized linear mixed model

KW - Overdispersion

KW - Serial correlation

UR - http://www.scopus.com/inward/record.url?scp=85054051567&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2018.09.002

DO - 10.1016/j.jspi.2018.09.002

M3 - Journal article

AN - SCOPUS:85054051567

VL - 200

SP - 20

EP - 31

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

ER -

ID: 203664686