A Delaporte Innovation Time Series Model for Dependent and Overdispersed Count Data
International Journal of Innovative Science and Research Technology · 02.03.2026 · 2158
DOI: 10.38124/ijisrt/26feb393 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovative Science and Research Technology, online erschienen am 02.03.2026. Verfasst von Enesi Latifat Oyiza und 4 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben.
Abstract (Originalsprache)
The Delaporte-DCINMA(q) model is a novel integer-valued moving average procedure for overdispersed count time series that is presented in this study. The model preserves discreteness through binomial thinning while overcoming the equidispersion limitation of Poisson-based models by utilizing Delaporte-distributed innovations. We establish the moment structures and important statistical features of the model. Simulation studies show the finite-sample performance and consistency of the estimator. The model's practical usefulness is confirmed by an application to U.S. polio death data, which effectively captures both considerable serial dependence and overdispersion. A versatile and reliable framework for examining correlated count data from a variety of disciplines is offered by the suggested model.
Zitieren
APA
Oyiza, E. L., Dolapo, S., Otaru, P., Onoghojobi, B., & R., B. O. (2026). A Delaporte Innovation Time Series Model for Dependent and Overdispersed Count Data. International Journal of Innovative Science and Research Technology, 2158. https://doi.org/10.38124/ijisrt/26feb393
Vancouver
Oyiza EL, Dolapo S, Otaru P, Onoghojobi B, R. BO. A Delaporte Innovation Time Series Model for Dependent and Overdispersed Count Data. International Journal of Innovative Science and Research Technology. 2026:2158. doi:10.38124/ijisrt/26feb393
BibTeX
@article{Oyiza2026,
author = {Oyiza, Enesi Latifat and Dolapo, Shobanke and Otaru, Paul and Onoghojobi, Benson and R., Babatunde O.},
title = {A Delaporte Innovation Time Series Model for Dependent and Overdispersed Count Data},
journal = {International Journal of Innovative Science and Research Technology},
year = {2026},
pages = {2158},
doi = {10.38124/ijisrt/26feb393}
}
EUVIMED