An alternative hyper-Poisson integer-valued GARCH model with application to polio, internet protocol and COVID-19 data

Kee Wah Fo; Seng Huat Ong; Choung Min Ng; You Beng Koh; Kee Wah Fo; Seng Huat Ong; Choung Min Ng; You Beng Koh

doi:10.3934/math.20231491

AIMS Mathematics

2023, Volume 8, Issue 12: 29116-29139. doi: 10.3934/math.20231491

Previous Article Next Article

Research article Special Issues

An alternative hyper-Poisson integer-valued GARCH model with application to polio, internet protocol and COVID-19 data

1.
Faculty of Engineering and Quantity Surveying, INTI International University, Persiaran Bandar Baru Nilai, 71800 Nilai, Malaysia
2.
Institute of Actuarial Science and Data Analytics, UCSI University, 56000 Kuala Lumpur, Malaysia
3.
Institute of Mathematical Sciences, Faculty of Science, Universiti Malaya, 50603 Kuala Lumpur, Malaysia

Received: 14 June 2023 Revised: 05 October 2023 Accepted: 17 October 2023 Published: 26 October 2023
MSC : 60G10, 62M10, 62F10

Time series of counts are observed widely in actuarial science, finance, epidemiology and biology. These time series may exhibit over-, equi- and under-dispersion. The Poisson distribution is commonly used in count time series models, but it is restricted by the equality of mean and variance. Other distributions such as the generalized Poisson, double Poisson, hyper-Poisson, and COM-Poisson distributions have been proposed to replace the Poisson distribution to model the different levels of dispersion in time series of counts. These models have certain limitations such as complex expressions for the mean and variance which complicate the formulation as GARCH models. In this study, we propose an alternative hyper-Poisson (AHP) distribution, with simple forms of conditional mean and variance, for an integer-valued GARCH (INGARCH) model for time series of counts that also exhibit the different levels of dispersion. We demonstrate that the AHP-INGARCH model is comparable to some existing INGARCH models. Additionally, the model can cover a wider range of dispersion. The maximum likelihood estimation can be used to estimate the parameters of the proposed model. Applications to three real-life data sets related to polio, internet protocol and daily COVID-19 new deaths underscore the usefulness of the proposed model in studying both over-dispersed and under-dispersed time series of counts.
- alternative hyper-Poisson,
- integer-valued GARCH,
- time series of counts,
- polio,
- internet protocol,
- COVID-19
Citation: Kee Wah Fo, Seng Huat Ong, Choung Min Ng, You Beng Koh. An alternative hyper-Poisson integer-valued GARCH model with application to polio, internet protocol and COVID-19 data[J]. AIMS Mathematics, 2023, 8(12): 29116-29139. doi: 10.3934/math.20231491

Related Papers:

Abstract

Time series of counts are observed widely in actuarial science, finance, epidemiology and biology. These time series may exhibit over-, equi- and under-dispersion. The Poisson distribution is commonly used in count time series models, but it is restricted by the equality of mean and variance. Other distributions such as the generalized Poisson, double Poisson, hyper-Poisson, and COM-Poisson distributions have been proposed to replace the Poisson distribution to model the different levels of dispersion in time series of counts. These models have certain limitations such as complex expressions for the mean and variance which complicate the formulation as GARCH models. In this study, we propose an alternative hyper-Poisson (AHP) distribution, with simple forms of conditional mean and variance, for an integer-valued GARCH (INGARCH) model for time series of counts that also exhibit the different levels of dispersion. We demonstrate that the AHP-INGARCH model is comparable to some existing INGARCH models. Additionally, the model can cover a wider range of dispersion. The maximum likelihood estimation can be used to estimate the parameters of the proposed model. Applications to three real-life data sets related to polio, internet protocol and daily COVID-19 new deaths underscore the usefulness of the proposed model in studying both over-dispersed and under-dispersed time series of counts.

References

[1]	G. E. Bardwell, E. L. Crow, A two-parameter family of hyper-Poisson distributions, J. Amer. Stat. Assoc., 59 (1964), 133–141. http://dx.doi.org/10.1080/01621459.1964.10480706 doi: 10.1080/01621459.1964.10480706
[2]	T. Bollerslev, Generalized autoregressive conditional heteroskedasticity, J. Economet., 31 (1986), 307–327. http://dx.doi.org/10.1016/0304-4076(86)90063-1 doi: 10.1016/0304-4076(86)90063-1
[3]	M. Bourguignon, C. H. Weiß, An INAR (1) process for modeling count time series with equidispersion, underdispersion and overdispersion, Test, 26 (2017), 847–868. http://dx.doi.org/10.1007/s11749-017-0536-4 doi: 10.1007/s11749-017-0536-4
[4]	M. Bourguignon, J. Rodrigues, M. Santos-Neto, Extended Poisson INAR (1) processes with equidispersion, underdispersion and overdispersion, J. Appl. Stat., 46 (2018), 101–118. http://dx.doi.org/10.1080/02664763.2018.1458216 doi: 10.1080/02664763.2018.1458216
[5]	A. C. Cameron, P. K. Trivedi, Regression analysis of count Data, 2 Eds., Cambridge University Press, New York, 2013. https://doi.org/10.1017/CBO9781139013567
[6]	P. C. Consul, Generalized Poisson distributions: Properties and applications, 1 Ed., Marcel Dekker, New York, 1989.
[7]	P. C. Consul, F. Famoye, Generalized Poisson regression model, Commun. Stat. -Theory Methods., 21 (1992), 89–109. http://dx.doi.org/10.1080/03610929208830766 doi: 10.1080/03610929208830766
[8]	R. W. Conway, W. L. Maxwell, A queuing model with state dependent service rates, J. Ind. Eng., 12 (1962), 132–136.
[9]	C. Czado, T. Gneiting, L. Held, Predictive model assessment for count data, Biometrics, 65 (2009), 1254–1261. https://dx.doi.org/10.1111/j.1541-0420.2009.01191.x doi: 10.1111/j.1541-0420.2009.01191.x
[10]	A. P. Dawid, Present position and potential developments: Some personal views: Statistical theory: The prequential approach, J. Roy. Stat. Soc. A, 147 (1984), 278–292. https://dx.doi.org/10.2307/2981683 doi: 10.2307/2981683
[11]	B. Efron, Double exponential families and their use in generalized linear regression, J. Amer. Stat Assoc., 81 (1986), 709–721. https://dx.doi.org/10.1080/01621459.1986.10478327 doi: 10.1080/01621459.1986.10478327
[12]	R. F. Engle, J. R. Russell, Autoregressive conditional duration: A new model for irregularly spaced transaction data, Econometrica, 66 (1998), 1127–1162. https://dx.doi.org/10.2307/2999632 doi: 10.2307/2999632
[13]	F. Famoye, Restricted generalized Poisson regression model, Commun. Stat. -Theory Methods., 22 (1993), 1335–1354. https://dx.doi.org/10.1080/03610929308831089 doi: 10.1080/03610929308831089
[14]	R. Ferland, A. Latour, D. Oraichi, Integer-valued GARCH processes. J. Time Ser. Anal., 27 (2006), 923–942. https://dx.doi.org/10.1111/j.1467-9892.2006.00496.x doi: 10.1111/j.1467-9892.2006.00496.x
[15]	K. Fokianos, A. Rahbek, D. Tjostheim, Poisson autoregression, J. Amer. Stat. Assoc., 104 (2009), 1430–1439. https://dx.doi.org/10.2139/ssrn.1323291 doi: 10.2139/ssrn.1323291
[16]	K. Fokianos, Some recent progress in count time series, Statistics, 45 (2011), 49–58. https://dx.doi.org/10.1080/02331888.2010.541250 doi: 10.1080/02331888.2010.541250
[17]	E. Gómez-Déniz, J. M. Sarabia, E. Calderín-Ojeda, A new discrete distribution with actuarial applications, Insur. Math. Econ., 48 (2011), 406–412. https://doi.org/10.1016/j.insmatheco.2011.01.007 doi: 10.1016/j.insmatheco.2011.01.007
[18]	S. D. Guikema, J. P. Goffelt, A flexible count data regression model for risk analysis, Risk Anal, 28 (2008), 213–223. https://dx.doi.org/10.1111/j.1539-6924.2008.01014.x doi: 10.1111/j.1539-6924.2008.01014.x
[19]	A. Heinen, Modelling time series count data: An autoregressive conditional Poisson model, SSRN E-journal, 2003. https://dx.doi.org/10.2139/ssrn.1117187
[20]	N. L. Johnson, A. W. Kemp, S. Kotz, Univariate discrete distributions, 3 Eds., John Wiley & Sons, New Jersey, 2005.
[21]	R. C. Jung, M. Kukuk, R. Liesenfeld, Time series of count data: Modeling, estimation and diagnostics, Comput. Stat. Data. Anal., 51 (2006), 2350–2364. https://dx.doi.org/10.1016/j.csda.2006.08.001 doi: 10.1016/j.csda.2006.08.001
[22]	R. C. Jung, A. Tremayne, Useful models for time series of counts or simply wrong ones? AStA. Adv. Stat. Anal., 95 (2011), 59–91. https://dx.doi.org/10.1007/s10182-010-0139-9 doi: 10.1007/s10182-010-0139-9
[23]	Y. Kang, D. Wang, K. Yang, Y. Zhang, A new thinning-based INAR (1) process for underdispersed or overdispersed counts. J. Korean Stat. Soc., 49 (2020), 324–349. https://doi.org/10.1007/s42952-019-00010-2 doi: 10.1007/s42952-019-00010-2
[24]	Y. B. Koh, N. A. Bukhari, I. Mohamed, Parameter-driven state-space model for integer-valued time series with application, J. Stat. Comput. Simul., 89 (2019), 1394–1409. https://doi.org/10.1080/00949655.2019.1582653 doi: 10.1080/00949655.2019.1582653
[25]	C. S. Kumar, B. U. Nair, An alternative hyper-Poisson distribution, Statistica, 72 (2012), 357–369. https://dx.doi.org/10.6092/issn.1973-2201/3652 doi: 10.6092/issn.1973-2201/3652
[26]	D. Lord, S. D. Guikema, S. R. Geedipally, Application of the Conway-Maxwell-Poisson generalized linear model for analyzing motor vehicle crashes, Accid. Anal. Prev., 40 (2008), 1123–1134. https://dx.doi.org/10.1016/j.aap.2007.12.003 doi: 10.1016/j.aap.2007.12.003
[27]	L. Qian, F. Zhu, A flexible model for time series of counts with overdispersion or underdispersion, zero-inflation and heavy-tailedness, Commun. Math. Stat., (2023). https://doi.org/10.1007/s40304-022-00327-1
[28]	A. J. Sáez-Castillo, A. Conde-Sánchez, A hyper-Poisson regression model for overdispersed and underdispersed count data, Comput. Stat. Data Anal., 61 (2013), 148–157. https://dx.doi.org/10.1016/j.csda.2012.12.009 doi: 10.1016/j.csda.2012.12.009
[29]	G. Shmueli, T. P. Minka, J. B. Kadane, S. Borle, P. Boatwright, A useful distribution for fitting discrete data: revival of the Conway–Maxwell–Poisson distribution, J. Roy. Stat. Soc. C, 54 (2005), 127–142. https://dx.doi.org/10.1111/j.1467-9876.2005.00474.x doi: 10.1111/j.1467-9876.2005.00474.x
[30]	C. H. Weiß, Controlling correlated processes of Poisson counts, Qual. Reliab. Eng. Int., 23 (2007), 741–754. https://dx.doi.org/10.1002/qre.875 doi: 10.1002/qre.875
[31]	C. H. Weiß, Modelling time series of counts with overdispersion, Stat. Method. Appl., 18 (2009), 507–519. https://dx.doi.org/10.1007/s10260-008-0108-6 doi: 10.1007/s10260-008-0108-6
[32]	C. H. Weiß, Thinning operations for modeling time series of counts—a survey, AStA Adv. Stat. Anal., 92 (2008), 319–341. https://dx.doi.org/10.1007/s10182-008-0072-3 doi: 10.1007/s10182-008-0072-3
[33]	C. H. Weiß, Integer-valued autoregressive models for counts showing underdispersion, J. Appl. Stat., 40 (2013), 1931–1948. https://doi.org/10.1080/02664763.2013.800034 doi: 10.1080/02664763.2013.800034
[34]	C. H. Weiß, An Introduction to Discrete-valued Times Series, 1 Ed., John Wiley & Sons Ltd, Hoboken, 2018. https://doi.org/10.1002/9781119097013
[35]	H. White, Maximum likelihood estimation of misspecified models, Econometrica, 50 (1982), 1–25. https://dx.doi.org/10.2307/1912526 doi: 10.2307/1912526
[36]	R. Winkelmann, Econometric analysis of count data, 5 Eds., Springer Science & Business Media, New York, 2008.
[37]	K. Yang, K. Yao, D. Wang, H. Li, Y. Diao, Modeling overdispersed or underdispersed count data with generalized Poisson integer-valued autoregressive processes, Metrika, 82 (2019), 863–889. https://doi.org/10.1007/s00184-019-00714-9 doi: 10.1007/s00184-019-00714-9
[38]	S. L. Zeger, A regression model for time series of counts, Biometrika, 75 (1988), 621–629. https://dx.doi.org/10.2307/2336303 doi: 10.2307/2336303
[39]	F. Zhu, A negative binomial integer-valued GARCH model, J. Time Ser. Anal., 32 (2011), 54–67. https://dx.doi.org/10.1111/j.1467-9892.2010.00684.x doi: 10.1111/j.1467-9892.2010.00684.x
[40]	F. Zhu, Modeling time series of counts with COM-Poisson INGARCH models, Math. Comput. Model., 56 (2012), 191–203. https://dx.doi.org/10.1016/j.mcm.2011.11.069 doi: 10.1016/j.mcm.2011.11.069
[41]	F. Zhu, Modeling overdispersed or underdispersed count data with generalized Poisson integer-valued GARCH models, J. Math. Anal. Appl., 389 (2012), 58–71. https://dx.doi.org/10.1016/j.jmaa.2011.11.042 doi: 10.1016/j.jmaa.2011.11.042
[42]	Y. Zou, S. R. Geedipally, D. Lord, Evaluating the double Poisson generalized linear model, Accid. Anal. Prev., 59 (2013), 497–505. https://dx.doi.org/10.1016/j.aap.2013.07.017 doi: 10.1016/j.aap.2013.07.017

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)