A study on the integer-valued time series models with overdispersion

• Main Authors: Pei-Hsun Tsai, 蔡沛洵
• Other Authors: Mei-Hui Guo
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/05765042285169986771

碩士 === 國立中山大學 === 應用數學系研究所 === 101 === Time series of counts observed in practice often exhibit overdispersion. The integer-valued generalized autoregressive conditional heteroscedastic (Ingarch) models are commonly used for count time series with overdispersion. We assume the conditional mean of an Ingarch model follows a Poisson distribution or other distributions, such as the negative Binomial distribution or the Generalized Poisson distribution. In this study, we investigate the properties, estimation of these Ingarch models. Two estimation methods: conditional least squares and maximum likelihood approach are considered. Numerical studies are performed to compare the properties and estimation of these Ingarch models.