详细信息
非对称三参数Student-t分布的参数估计及应用
Parameter Estimation and Application of Asymmetric Three-parameter Student-t Distribution
文献类型:期刊文献
中文题名:非对称三参数Student-t分布的参数估计及应用
英文题名:Parameter Estimation and Application of Asymmetric Three-parameter Student-t Distribution
作者:郝家岗[1];钱夕元[1]
机构:[1]华东理工大学理学院,上海200237
年份:2021
期号:11
起止页码:25
中文期刊名:统计与决策
外文期刊名:Statistics & Decision
收录:CSTPCD;;国家哲学社会科学学术期刊数据库;北大核心:【北大核心2020】;CSSCI:【CSSCI2021_2022】;
基金:国家高科技研究发展计划(“863计划”)资助项目(2015AA20107)。
语种:中文
中文关键词:偏态厚尾;非对称Student-t分布;贝叶斯估计;数据拟合
外文关键词:skew fat tail;asymmetric Student-t distribution;Bayesian estimation;data fitting
摘要:文章基于t分布构造了一个新的非对称三参数Student-t分布。推导了新分布的累计分布函数、分位数函数、随机变量表达式和原点矩估计等基础性质,使用了矩方法、极大似然估计法和贝叶斯估计法等进行了参数估计,并通过生成模拟数据的方法比较验证了三种方法的合理性。最后,将新分布代入实例中拟合,结果表明新分布相比其他的分布,在非对称和厚尾方面具有更好的拟合能力。
This paper constructs a new asymmetric three-parameter Student-t distribution based on t distribution, deduces the basic properties of the new distribution, such as the cumulative distribution function, the quantile function, the expression of random variable and the estimation of the origin moment, and uses the moment method, maximum likelihood estimation method and Bayesian estimation method to estimate the parameters. Then, the paper compares and verifies the rationality of the three methods by generating simulation data. Finally, the paper substitutes the new distribution into the example for fitting, and the results show that the new distribution has better fitting ability than other distributions in terms of asymmetry and fat tail.
参考文献:
正在载入数据...
