Springer New York;Berlin/Heidelberg: Springer-Verlag
摘要:
摘要： Typhoon is one of the most destructive disasters in Taiwan, which usually causes many floods and mudslides and prevents the electrical and water supply. Prior to its arrival, how to accurately forecast the path and rainfall of typhoon are important issues. In the past, a regression-based model was the most applied statistical method to evaluate the associated problems. However, it generally ignored the spatial dependence in the data, resulting in less accurate estimation and prediction, and the importance of particular explanatory variables may not be apparent. Therefore, in this paper we focus on assessing the spatial risk variations regarding the typhoon cumulated rainfall at Taipei with respect to typhoon locations by using the spatial hierarchical Bayesian model combined with the spatial conditional autoregressive model, where the model parameters are estimated by designing a family of stochastic algorithms based on a Markov chain Monte Carlo technique. The proposed method is applied to a real data set of Taiwan for illustration. Also, some important explanatory variables regarding the typhoon cumulated rainfall at Taipei are indicated as well. 其他題名： Stoch Environ Res Risk Assess 出版者： Berlin/Heidelberg: Springer-Verlag 出版日期： 2012-05-01 出處： Stochastic environmental research and risk assessment, 2012-05, Vol.26 (4), p.509-517 資源來源： Springer Nature Link 版權： Springer-Verlag 2011 版權： Springer-Verlag 2012 識別號： ISSN: 1436-3240 識別號： EISSN: 1436-3259 識別號： DOI: 10.1007/s00477-011-0508-2
