摘要
为了拟合不同概率分布,需要对GLD分布的有效区域进行讨论,将平面划分为8个不同区域分别来探讨GLD分布有效区域问题。采用百分位法来拟合已知分布,并检验拟合结果是否在有效区域内。在拟合算法上采用matlab编写程序。给出了卡方分布和对数分布具体地拟合过程。
In order to fit different probability distributions, the effective regions of the GLD distribution need to be discussed. The plane is divided into eight different regions to explore the GLD effective region problems respectively. The percentile method is used to fit the known distribution and to check if the fitted result is within the scope of the effective region. Matlab programming is introduced in the fitting algorithm. The specific fitting process of the chi-square distribution and the of logarithmic distribution are given.
引文
[1]Lakhany A,Mausser H.Estimating the parameters of the generalized lambda distribution[J].Algo Research Quarterly,2000,3(3):47-58.
[2]Karian Z A,Dudewicz E J.Handbook of Fitting Satatistica Distributions with R[M].New York:CRC Press,2000:20-25.
[3]Karian Z A,Dudewicz E J.Handbook of Fitting Satatistica Distributions with R[M].New York:CRC Press,2000:171-174.
[4]Fournier B,Rupin N,Bigereller M,et al.Estimating the parameters of generalized lambda distribution[J].Computational Statistics&Data Analysis,2007,251:2813-2835.
[5]陈刚,王梦婕.卡方分布密度函数与分布函数的渐近展开[J].南京师大学报,2014,37(3):39-43.
[6]李树有,徐美进,刘秀娟.应用数理统计[M].沈阳:东北大学出版社,2015.