摘要
基于马尔科夫链蒙特卡洛(简记为MCMC)模拟的参数贝叶斯估计,对改进的广义帕累托分布(简记为MGPD)模型进行了优化,并利用该模型得到了地质灾害损失的在险损失值(简记为VaR)和条件损失值(简记为CVaR).以湖南娄底市地质灾害损失数据进行实证分析及模型适应性检验,结果表明:优化后的模型不仅具有很好的极值数据描述能力,而且具有较强的适用性.
We used Bayesian estimation based on Markov Chain simulation to optimize the meliorated Generalized Pareto Distribution model (MGPD), and obtained the estimation of the Value at Risk(VaR) and Conditionl Value at Risk(CVaR). The empirical study and adaptability test of the model were based on geological disasters loss data of Loudi City in Hunan Prov- ince. The conclusion shows the optimized model has not only excellent ability in describing the data, but also extensive applica- bility.
出处
《经济数学》
2015年第2期101-106,共6页
Journal of Quantitative Economics
基金
国家自然科学基金项目(11171044)
湖南省国土资源科技项目(2013-28)资助
关键词
马尔科夫链蒙特卡洛模拟
贝叶斯估计
改进广义帕累托分布
地质灾害
Markov Chain Monte Carlo simulation
Bayesian estimations meliorated generalized Pareto distribution model
geological disaster