Models for autoregressive processes of bounded counts: how different are they? (English) Zbl 1505.62225
Summary: We focus on purely autoregressive (AR)-type models defined on the bounded range \(\{0,1,\dots,n\}\) with a fixed upper limit \(n\in \mathbb{N}\). These include the binomial AR model, binomial AR conditional heteroscedasticity (ARCH) model, binomial-variation AR model with their linear conditional mean, nonlinear max-binomial AR model, and binomial logit-ARCH model. We consider the key problem of identifying which of these AR-type models is the true data-generating process. Despite the volume of the literature on model selection, little is known about this procedure in the context of nonnested and nonlinear time series models for counts. We consider the most popular approaches used for model identification, Akaike’s information criterion and the Bayesian information criterion, and compare them using extensive Monte Carlo simulations. Furthermore, we investigate the properties of the fitted models (both the correct and wrong models) obtained using maximum likelihood estimation. A real-data example demonstrates our findings.
MSC:
| 62-08 | Computational methods for problems pertaining to statistics |
| 62M10 | Time series, auto-correlation, regression, etc. in statistics (GARCH) |
Keywords:
binomial autoregressive models; count time series; model adequacy; model selection; parameter estimationReferences:
| [1] | Akaike, H., A new look at the statistical model identification, IEEE Trans Autom Control, 19, 6, 717-723 (1974) · Zbl 0314.62039 |
| [2] | Al-Osh, MA; Alzaid, AA, First-order integer-valued autoregressive (INAR(1)) process, J Time Ser Anal, 8, 3, 261-275 (1987) · Zbl 0617.62096 |
| [3] | Burnham, KP; Anderson, DR, Model selection and multimodel inference: a practical information-theoretic approach (2002), New York: Springer, New York · Zbl 1005.62007 |
| [4] | Cavanaugh, JE; Neath, AA, The Akaike information criterion: background, derivation, properties, application, interpretation, and refinements, WIREs Comput Stat, 11, 3, e1460 (2019) · Zbl 07909157 |
| [5] | Chen H, Li Q, Zhu F (2019) Two classes of dynamic binomial integer-valued ARCH models. Braz J Probab Stat, forthcoming |
| [6] | Claeskens, G.; Hjort, NL, Model selection and model averaging (2008), Cambridge: Cambridge University Press, Cambridge · Zbl 1166.62001 |
| [7] | Cui, Y.; Lund, R., A new look at time series of counts, Biometrika, 96, 4, 781-792 (2009) · Zbl 1179.62116 |
| [8] | Davis, RA; Resnick, SI, Basic properties and prediction of max-ARMA processes, Adv Appl Prob, 21, 4, 781-803 (1989) · Zbl 0716.62098 |
| [9] | Diop ML, Kenge W (2020) Consistent model selection procedure for general integer-valued time series. arXiv:2002.08789 |
| [10] | Dziak JJ, Coffman DL, Lanza ST, Li R, Jermiin LS (2019) Sensitivity and specificity of information criteria. Brief Bioinform, bbz016 |
| [11] | Gouveia, S.; Möller, TA; Weiß, CH; Scotto, MG, A full ARMA model for counts with bounded support and its application to rainy-days time series, Stoch Environ Res Risk Assess, 32, 9, 2495-2514 (2018) |
| [12] | Grunwald, G.; Hyndman, RJ; Tedesco, L.; Tweedie, RL, Non-Gaussian conditional linear AR(1) models, Aust N Z J Stat, 42, 4, 479-495 (2000) · Zbl 1018.62065 |
| [13] | Jung, RC; McCabe, BPM; Tremayne, AR; Davis, RA, Model validation and diagnostics, Handbook of discrete-valued time series, 189-218 (2016), Boca Raton: CRC Press, Boca Raton |
| [14] | Katz, RW, On some criteria for estimating the order of a Markov chain, Technometrics, 23, 3, 243-249 (1981) · Zbl 0485.62086 |
| [15] | Kunihama, T.; Omori, Y.; Zhang, Z., Efficient estimation and particle filter for max-stable processes, J Time Ser Anal, 33, 1, 61-80 (2012) · Zbl 1300.62094 |
| [16] | McKenzie, E., Some simple models for discrete variate time series, Water Res Bull, 21, 4, 645-650 (1985) |
| [17] | Neath, AA; Cavanaugh, JE, The Bayesian information criterion: background, derivation, and applications, WIREs Comput Stat, 4, 2, 199-203 (2012) |
| [18] | Psaradakis, Z.; Sola, M.; Spagnoloy, F., Selecting nonlinear time series models using information criteria, J Time Ser Anal, 30, 4, 369-394 (2009) · Zbl 1224.62067 |
| [19] | Rinke, S.; Sibbertsen, P., Information criteria for nonlinear time series models, Stud Nonlinear Dyn Econ, 20, 3, 325-341 (2016) · Zbl 1507.62294 |
| [20] | Ristić, MM; Weiß, CH; Janjić, AD, A binomial integer-valued ARCH model, Int J Biostat, 12, 2, 20150051 (2016) |
| [21] | Schwarz, G., Estimating the dimension of a model, Ann Stat, 6, 2, 461-464 (1978) · Zbl 0379.62005 |
| [22] | Steutel, FW; van Harn, K., Discrete analogues of self-decomposability and stability, Ann Probab, 7, 5, 893-899 (1979) · Zbl 0418.60020 |
| [23] | Weiß, CH, A new class of autoregressive models for time series of binomial counts, Commun Stat Theory Methods, 38, 4, 447-460 (2009) · Zbl 1159.62059 |
| [24] | Weiß, CH, An introduction to discrete-valued time series (2018), Chichester: Wiley, Chichester · Zbl 1407.62009 |
| [25] | Weiß, CH; Feld, MH-JM, On the performance of information criteria for model identification of count time series, Stud Nonlinear Dyn Econom, 24, 1, 20180012 (2020) · Zbl 07675514 |
| [26] | Weiß, CH; Feld, MH-JM; Mamode Khan, N.; Sunecher, Y., INARMA modeling of count time series, Stats, 2, 2, 284-320 (2019) |
| [27] | Weiß, CH; Kim, H-Y, Parameter estimation for binomial AR(1) models with applications in finance and industry, Stat Pap, 54, 3, 563-590 (2013) · Zbl 1307.62061 |
| [28] | Weiß, CH; Pollett, PK, Chain binomial models and binomial autoregressive processes, Biometrics, 68, 3, 815-824 (2012) · Zbl 1272.62099 |
| [29] | Weiß, CH; Scotto, MG; Möller, TA; Gouveia, S., The max-BARMA models for counts with bounded support, Stat Probab Lett, 143, 28-36 (2018) · Zbl 1414.62384 |
| [30] | Zhang, Z.; Smith, RL, On the estimation and application of max-stable processes, J Stat Plan Inference, 140, 5, 1135-1153 (2010) · Zbl 1181.62150 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
