First-order multivariate integer-valued autoregressive model with multivariate mixture distributions
Weiyang Yu,Haitao Zheng
DOI: https://doi.org/10.1080/00949655.2024.2364040
IF: 1.225
2024-06-17
Journal of Statistical Computation and Simulation
Abstract:The univariate integer-valued time series has been extensively studied, but literature on multivariate integer-valued time series models is quite limited and the complex correlation structure among the multivariate integer-valued time series is barely discussed. In this study, we proposed a first-order multivariate integer-valued autoregressive model to characterize the correlation among multivariate integer-valued time series with higher flexibility. Under the general conditions, we established the stationarity and ergodicity of the proposed model. With the proposed method, we discussed the models with multivariate Poisson-lognormal distribution and multivariate geometric-logitnormal distribution and the corresponding properties. The estimation method based on EM algorithm was developed for the model parameters and extensive simulation studies were performed to evaluate the effectiveness of proposed estimation method. Finally, a real crime dataset was analysed to demonstrate the advantage of the proposed model with comparison to the other models.
statistics & probability,computer science, interdisciplinary applications