文献类型：期刊文献

中文题名：基于泊松分布的森林火灾发生数贝叶斯估计模型

英文题名：Application of Bayesian Method in Forest Fire Occurrence Model Base on Poisson Distribution

作者：肖云丹[1] 纪平[1]

第一作者：肖云丹

机构：[1]中国林业科学研究院资源信息所

年份：2018

卷号：0

期号：12

起止页码：8-11

中文期刊名：林业科技通讯

外文期刊名：Forest Science and Technology

基金：中央级公益性科研院所基本科研业务费专项(CAFYBB2017QA010)

语种：中文

中文关键词：贝叶斯估计;森林火灾发生;气象因子;Poisson模型;零膨胀Poisson模型

外文关键词：Bayesian Method;forest fire;meteorological variables;Poisson regression model;zero- Poisson model

分类号：S223

摘要：对黔南区森林火灾发生数与气象因子进行分析,以Poisson和零膨胀Poisson为基础,通过贝叶斯方法建立黔南地区火险天气森林火灾预测模型。结果表明:零膨胀Poisson模型拟合效果比Poisson模型拟合好。同时还发现,利用贝叶斯法估计森林火灾发生模型能够很好地评价森林火灾发生模型的不确定性。
Climate change is a popular topic which affects the forest fire occurrence. It is important for forecasting forest fire to analyze the relationship between forest fire occurrence and climatic factors. However,there existed the uncertainty in the relationships. The uncertainties were most from the climatic factors and relevant parameters, and statistical model structures. In this study, based on the forest fire occurrence and meteorological variables in spring fireproofing period in Qiannan area, Poisson regression model and zero-Poisson model were established respectively to predict the forest fires by Bayesian Method. The results showed that ZIP model was better than Poisson model. Additionally, The model reliability using Bayesian Method was better than using classical method.