Indirect inference for spatio-temporal autoregression models
نویسندگان
چکیده
In this note we introduce a new inferential method for STAR (spatio-temporal autoregression) models. Due to the complexity of such models the maximum likelihood estimation is difficult to undertake when several nearest neighbours are included in the model, see Ali (1979). Moreover, only approximate likelihoods are available in practice because of the observations lying on the edges of the spatial domain. On the other hand, simpler estimation methods such as least squares and Yule-Walker are not generally consistent. With this background, we propose the use of an inferential method based on an auxiliary model whose parameters can be consistently estimated with Yule-Walker. From this estimation, consistent estimators for the parameters of the STAR model of interest are then retrieved with the help of simulated data. The method and its asymptotic theory are presented in Section 2. In Section 3 we illustrate its small sample properties with a limited Monte Carlo experiment. The note is concluded in Section 4.
منابع مشابه
Spatio-temporal analysis of diurnal air temperature parameterization in Weather Stations over Iran
Diurnal air temperature modeling is a beneficial experimental and mathematical approach which can be used in many fields related to Geosciences. The modeling and spatio-temporal analysis of air Diurnal Temperature Cycle (DTC) was conducted using data obtained from 105 synoptic stations in Iran during the years 2013-2014 for the first time; the key variable for controlling the cosine term i...
متن کاملمعرفی شبکه های عصبی پیمانه ای عمیق با ساختار فضایی-زمانی دوگانه جهت بهبود بازشناسی گفتار پیوسته فارسی
In this article, growable deep modular neural networks for continuous speech recognition are introduced. These networks can be grown to implement the spatio-temporal information of the frame sequences at their input layer as well as their labels at the output layer at the same time. The trained neural network with such double spatio-temporal association structure can learn the phonetic sequence...
متن کاملFitting dynamic models using integrated nested Laplace approximations - INLA
In this paper we propose a computational framework to perform approximate Bayesian inference in linear and generalized linear dynamic models based on the Integrated Nested Laplace Approximation (INLA) approach, which overcomes some limitations of computational tools presently available in the dynamic modeling literature. We show how to formulate specific latent models in a state-space form, eve...
متن کاملAssessment of Neonate's Congenital Hypothyroidism Pattern Using Poisson Spatio-temporal Model in Disease Mapping under the Bayesian Paradigm during 2011-18 in Guilan, Iran
Background: Congenital Hypothyroidism (CH) is one of the reasons for mental retardation and defective growth in neonates. It can be treated if it is diagnosed early. The congenital hypothyroidism can be diagnosed using newborn screening in the first days after birth. Disease mapping helps to identify high-risk areas of the disease. This study aimed to evaluate the pattern of CH using the Poisso...
متن کاملEvaluation of Tests for Separability and Symmetry of Spatio-temporal Covariance Function
In recent years, some investigations have been carried out to examine the assumptions like stationarity, symmetry and separability of spatio-temporal covariance function which would considerably simplify fitting a valid covariance model to the data by parametric and nonparametric methods. In this article, assuming a Gaussian random field, we consider the likelihood ratio separability test, a va...
متن کامل