An EM algorithm for estimation of the parameters of the geometric minification INAR model
- 20 March 2022
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 92 (14) , 2941-2955
Abstract
One of the most important problems that arise when estimating parameters by the maximum likelihood method in INAR models, including minification models, is that the estimates cannot be presented in analytical form, but some numerical method must be used to find them. To solve this problem, we present an EM algorithm. As this problem is difficult to solve based on the original definition of the model, in this manuscript, we first show that the model can be presented in an equivalent form. Then, based on that equivalent form, we construct an EM algorithm for estimating the parameters of the model. Finally, the quality of the estimates and the speed of the algorithm were observed on the simulated data.Keywords
Funding Information
- Ministry of Education, Science and Technological Development of the Republic of Serbia (451-03-9/2021-14/200124)
This publication has 5 references indexed in Scilit:
- A geometric minification integer-valued autoregressive modelApplied Mathematical Modelling, 2021
- A negative binomial thinning‐based bivariate INAR(1) processStatistica Neerlandica, 2020
- Modeling time series of counts with a new class of INAR(1) modelStatistical Papers, 2015
- Studying the effect of weather conditions on daily crash counts using a discrete time-series modelAccident Analysis & Prevention, 2008
- Maximum Likelihood from Incomplete Data Via the EM AlgorithmJournal of the Royal Statistical Society Series B: Statistical Methodology, 1977