Home > Research > Publications & Outputs > Log-linear Poisson autoregression

Links

Text available via DOI:

View graph of relations

Log-linear Poisson autoregression

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Log-linear Poisson autoregression. / Fokianos, K.; Tjøstheim, D.
In: Journal of Multivariate Analysis, Vol. 102, No. 3, 03.2011, p. 563-578.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Fokianos, K & Tjøstheim, D 2011, 'Log-linear Poisson autoregression', Journal of Multivariate Analysis, vol. 102, no. 3, pp. 563-578. https://doi.org/10.1016/j.jmva.2010.11.002

APA

Fokianos, K., & Tjøstheim, D. (2011). Log-linear Poisson autoregression. Journal of Multivariate Analysis, 102(3), 563-578. https://doi.org/10.1016/j.jmva.2010.11.002

Vancouver

Fokianos K, Tjøstheim D. Log-linear Poisson autoregression. Journal of Multivariate Analysis. 2011 Mar;102(3):563-578. Epub 2010 Nov 25. doi: 10.1016/j.jmva.2010.11.002

Author

Fokianos, K. ; Tjøstheim, D. / Log-linear Poisson autoregression. In: Journal of Multivariate Analysis. 2011 ; Vol. 102, No. 3. pp. 563-578.

Bibtex

@article{b237aaddde064c049bc5f2e0ddd56d49,
title = "Log-linear Poisson autoregression",
abstract = "We consider a log-linear model for time series of counts. This type of model provides a framework where both negative and positive association can be taken into account. In addition time dependent covariates are accommodated in a straightforward way. We study its probabilistic properties and maximum likelihood estimation. It is shown that a perturbed version of the process is geometrically ergodic, and, under some conditions, it approaches the non-perturbed version. In addition, it is proved that the maximum likelihood estimator of the vector of unknown parameters is asymptotically normal with a covariance matrix that can be consistently estimated. The results are based on minimal assumptions and can be extended to the case of log-linear regression with continuous exogenous variables. The theory is applied to aggregated financial transaction time series. In particular, we discover positive association between the number of transactions and the volatility process of a certain stock.",
author = "K. Fokianos and D. Tj{\o}stheim",
year = "2011",
month = mar,
doi = "10.1016/j.jmva.2010.11.002",
language = "English",
volume = "102",
pages = "563--578",
journal = "Journal of Multivariate Analysis",
issn = "0047-259X",
publisher = "Academic Press Inc.",
number = "3",

}

RIS

TY - JOUR

T1 - Log-linear Poisson autoregression

AU - Fokianos, K.

AU - Tjøstheim, D.

PY - 2011/3

Y1 - 2011/3

N2 - We consider a log-linear model for time series of counts. This type of model provides a framework where both negative and positive association can be taken into account. In addition time dependent covariates are accommodated in a straightforward way. We study its probabilistic properties and maximum likelihood estimation. It is shown that a perturbed version of the process is geometrically ergodic, and, under some conditions, it approaches the non-perturbed version. In addition, it is proved that the maximum likelihood estimator of the vector of unknown parameters is asymptotically normal with a covariance matrix that can be consistently estimated. The results are based on minimal assumptions and can be extended to the case of log-linear regression with continuous exogenous variables. The theory is applied to aggregated financial transaction time series. In particular, we discover positive association between the number of transactions and the volatility process of a certain stock.

AB - We consider a log-linear model for time series of counts. This type of model provides a framework where both negative and positive association can be taken into account. In addition time dependent covariates are accommodated in a straightforward way. We study its probabilistic properties and maximum likelihood estimation. It is shown that a perturbed version of the process is geometrically ergodic, and, under some conditions, it approaches the non-perturbed version. In addition, it is proved that the maximum likelihood estimator of the vector of unknown parameters is asymptotically normal with a covariance matrix that can be consistently estimated. The results are based on minimal assumptions and can be extended to the case of log-linear regression with continuous exogenous variables. The theory is applied to aggregated financial transaction time series. In particular, we discover positive association between the number of transactions and the volatility process of a certain stock.

U2 - 10.1016/j.jmva.2010.11.002

DO - 10.1016/j.jmva.2010.11.002

M3 - Journal article

VL - 102

SP - 563

EP - 578

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

SN - 0047-259X

IS - 3

ER -