Previous |  Up |  Next

Article

Title: Modeling a flexible uni/bimodal $\mathbb {Z}$-valued time series with discrete double Lindley innovations: Forecasting via hybrid and statistical approaches (English)
Author: Bakouch, Hassan S.
Author: Gabr, Mahmoud Mohamed
Author: El-Hadidy, Shimaa M.
Language: English
Journal: Applications of Mathematics
ISSN: 0862-7940 (print)
ISSN: 1572-9109 (online)
Volume: 71
Issue: 5
Year: 2026
Pages: 711-741
Summary lang: English
.
Category: math
.
Summary: The motivation for implementing this model arises from the need to analyze dependent count data that can take both positive and negative values, exhibit fluctuations around a zero mean, and demonstrate unimodal-bimodal aspects. To capture these features commonly observed in real-world applications, we propose a first-order integer-valued autoregressive process, INAR(1), with discrete double Lindley innovations. The model is constructed using a signed thinning mechanism, referred to as the relative binomial thinning operator, which extends the classical binomial thinning operator. The proposed model is appropriate for $\mathbb {Z}$-valued time series with both positive and negative dependence. The main statistical properties of the process are derived. Parameter estimation is addressed using modified conditional least squares, Yule-Walker, and conditional maximum likelihood methods, whose performance is evaluated through simulation studies. The practical relevance of the model is demonstrated using data from the Saudi stock market. Model adequacy is assessed via standardized residuals and the cumulative periodogram, and predictive performance is evaluated using a hybrid framework combining Support Vector Regression (SVR) with the proposed model, in comparison with bootstrap and classical approaches. (English)
Keyword: $\mathbb {Z}$-valued time series
Keyword: signed thinning operator
Keyword: statistical distribution on $\mathbb {Z}$
Keyword: simulation
Keyword: SVR-RBINAR-DDL(1)
Keyword: standardized Pearson residual
Keyword: bootstrap
Keyword: forecasting
MSC: 60G10
MSC: 62M10
MSC: 62M15
MSC: 62M20
MSC: 68T05
DOI: 10.21136/AM.2026.0070-26
.
Date available: 2026-10-12T13:00:07Z
Last updated: 2026-10-19
Stable URL: http://hdl.handle.net/10338.dmlcz/153800
.
Reference: [1] Ali, N. S. M., Mohammed, F. A.: The use of ARIMA, ANN and SVR models in time series hybridization with practical application.Int. J. Nonlinear Anal. Appl. 14 (2023), 87-102. 10.22075/ijnaa.2022.7168
Reference: [2] Al-Osh, M. A., Alzaid, A. A.: First-order integer-valued autoregressive (INAR(1)) process.J. Time Ser. Anal. 8 (1987), 261-275. Zbl 0617.62096, MR 0903755, 10.1111/j.1467-9892.1987.tb00438.x
Reference: [3] Alzaid, A. A., Omair, M. A.: Poisson difference integer valued autoregressive model of order one.Bull. Malays. Math. Sci. Soc. (2) 37 (2014), 465-485. Zbl 1291.62159, MR 3188051
Reference: [4] Barreto-Souza, W., Bourguignon, M.: A skew INAR(1) process on $\Bbb{Z}$.AStA, Adv. Stat. Anal. 99 (2015), 189-208. Zbl 1443.62242, MR 3328160, 10.1007/s10182-014-0236-2
Reference: [5] Bourguignon, M., Vasconcellos, K. L. P.: A new skew integer valued time series process.Stat. Methodol. 31 (2016), 8-19. Zbl 1487.62105, MR 3477725, 10.1016/j.stamet.2016.01.002
Reference: [6] Bu, R., McCabe, B.: Model selection, estimation and forecasting in INAR$(p)$ models: A likelihood-based Markov Chain approach.Int. J. Forecast. 24 (2008), 151-162. 10.1016/j.ijforecast.2007.11.002
Reference: [7] Chen, H., Zhang, J., Zhu, F.: A trinomial difference autoregressive model and its applications.Stat 12 (2023), Article ID e596, 16 pages. Zbl 07858720, MR 4641217, 10.1002/sta4.596
Reference: [8] Chesneau, C., Bakouch, H. S., Tomy, L., Veena, G.: A new discrete distribution on integers: Analytical and applied study on stock exchange and flood data.J. Stat. Manag. Syst. 25 (2022), 1899-1917. 10.1080/09720510.2021.1995199
Reference: [9] Chesneau, C., Kachour, M.: A parametric study for the first-order signed integer-valued autoregressive process.J. Stat. Theory Pract. 6 (2012), 760-782. Zbl 1425.62124, MR 3196577, 10.1080/15598608.2012.719816
Reference: [10] Cunha, E. T. da, Vasconcellos, K. L. P., Bourguignon, M.: A skew integer-valued time-series process with generalized Poisson difference marginal distribution.J. Stat. Theory Pract. 12 (2018), 718-743. MR 3878055, 10.1080/15598608.2018.1470046
Reference: [11] Davis, R. A., Fokianos, K., Holan, S. H., Joe, H., Livsey, J., Lund, R., Pipiras, V., Ravishanker, N.: Count time series: A methodological review.J. Am. Stat. Assoc. 116 (2021), 1533-1547. Zbl 1510.62356, MR 4309291, 10.1080/01621459.2021.1904957
Reference: [12] Guerrero, M. B., Barreto-Souza, W., Ombao, H.: Integer-valued autoregressive processes with prespecified marginal and innovation distributions: A novel perspective.Stoch. Models 38 (2022), 70-90. Zbl 1493.62525, MR 4359301, 10.1080/15326349.2021.1977141
Reference: [13] Hu, X.: Support Vector Machine and Its Application to Regression and Classification: Master's Thesis.Missouri State University, Springfield (2017).
Reference: [14] Hyndman, R. J., Koehler, A. B.: Another look at measures of forecast accuracy.Int. J. Forecast. 22 (2006), 679-688. 10.1016/j.ijforecast.2006.03.001
Reference: [15] Inusah, S., Kozubowski, T. J.: A discrete analogue of the Laplace distribution.J. Stat. Plann. Inference 136 (2006), 1090-1102. Zbl 1081.60011, MR 2181990, 10.1016/j.jspi.2004.08.014
Reference: [16] Jung, R. C., Tremayne, A. R.: Coherent forecasting in integer time series models.Int. J. Forecast. 22 (2006), 223-238. 10.1016/j.ijforecast.2005.07.001
Reference: [17] Jung, R. C., Tremayne, A. R.: Convolution-closed models for count time series with applications.J. Time Ser. Anal. 32 (2011), 268-280. Zbl 1290.62078, MR 2808254, 10.1111/j.1467-9892.2010.00697.x
Reference: [18] Kachour, M.: On the rounded integer-valued autoregressive process.Commun. Stat., Theory Methods 43 (2014), 355-376. Zbl 1462.62537, MR 3171042, 10.1080/03610926.2012.661506
Reference: [19] Kachour, M., Bakouch, H. S., Mohammadi, Z.: A new INAR(1) model for $\Bbb{Z}$-valued time series using the relative binomial thinning operator.Jahrb. Nationalökon. Stat. 243 (2023), 125-152. 10.1515/jbnst-2022-0059
Reference: [20] Kachour, M., Truquet, L.: A $p$-order signed integer-valued autoregressive (SINAR$(p)$) model.J. Time Ser. Anal. 32 (2011), 223-236. Zbl 1290.62079, MR 2808251, 10.1111/j.1467-9892.2010.00694.x
Reference: [21] Kachour, M., Yao, J. F.: First-order rounded integer-valued autoregressive (RINAR(1)) process.J. Time Ser. Anal. 30 (2009), 417-448. Zbl 1224.62060, MR 2536061, 10.1111/j.1467-9892.2009.00620.x
Reference: [22] Karlsen, H., Tjøstheim, D.: Consistent estimates for the NEAR(2) and NLAR(2) times series models.J. R. Stat. Soc., Ser. B 50 (1988), 313-320. MR 0964184, 10.1111/j.2517-6161.1988.tb01730.x
Reference: [23] Khoo, W. C., Ong, S. H., Atanu, B.: Coherent forecasting for a mixed integer-valued time series model.Mathematics 10 (2022), Article ID 2961, 15 pages. 10.3390/math10162961
Reference: [24] Kim, H.-Y., Park, Y.: A non-stationary integer-valued autoregressive model.Stat. Pap. 49 (2008), 485-502. Zbl 1148.62074, MR 2399216, 10.1007/s00362-006-0028-1
Reference: [25] Kiwon, F.: Frequentist Model Averaging for $\varepsilon$-Support Vector Regression: Ph.D. Thesis.McMaster University, Hamilton (2019).
Reference: [26] Lee, S., Lee, S.: Change point test for the conditional mean of time series of counts based on support vector regression.Entropy 23 (2021), Article ID 433, 17 pages. MR 4252529, 10.3390/e23040433
Reference: [27] McCabe, B. P. M., Martin, G. M.: Bayesian predictions of low count time series.Int. J. Forecast. 21 (2005), 315-330. 10.1016/j.ijforecast.2004.11.001
Reference: [28] McKenzie, E.: Some simple models for discrete variate time series.JAWRA, J. Am. Water Resour. Assoc. 21 (1985), 645-650. 10.1111/j.1752-1688.1985.tb05379.x
Reference: [29] Pascual, L., Romo, J., Ruiz, E.: Bootstrap predictive inference for ARIMA processes.J. Time Ser. Anal. 25 (2004), 449-465. Zbl 1062.62199, MR 2073232, 10.1111/j.1467-9892.2004.01713.x
Reference: [30] Scotto, M. G., Weià , C. H., Gouveia, S.: Thinning-based models in the analysis of integer-valued time series: A review.Stat. Model. 15 (2015), 590-618. Zbl 07259004, MR 3441230, 10.1177/1471082X15584701
Reference: [31] Steutel, F. W., Harn, K. van: Discrete analogues of self-decomposability and stability.Ann. Probab. 7 (1979), 893-899. Zbl 0418.60020, MR 0542141, 10.1214/aop/1176994950
Reference: [32] Weiß, C. H.: Thinning operations for modeling time series of counts: A survey.AStA, Adv. Stat. Anal. 92 (2008), 319-341. Zbl 1477.62256, MR 2426093, 10.1007/s10182-008-0072-3
Reference: [33] Yan, H., Wang, D.: Random environment $\Bbb{Z}$-valued AR(1) process with Poisson difference marginal distributions.(to appear) in Commun. Stat., Simul. Comput. 10.1080/03610918.2026.2613953
Reference: [34] Zhang, H., Wang, D., Zhu, F.: Inference for INAR$(p)$ processes with signed generalized power series thinning operator.J. Stat. Plan. Inference 140 (2010), 667-683. Zbl 1177.62110, MR 2558396, 10.1016/j.jspi.2009.08.012
Reference: [35] Zhang, H., Wang, D., Zhu, F.: Generalized RCINAR(1) process with signed thinning operator.Commun. Stat., Theory Methods 41 (2012), 1750-1770. Zbl 1270.62125, MR 2913009, 10.1080/03610926.2010.551452
.

Fulltext not available (moving wall 24 months)

Partner of
EuDML logo