通过对数变换给出了求两参数Birnbaum-Saunders(BS)疲劳寿命分布BS(α,β)在全样本场合下参数的对数矩估计,并通过大量Monte-Carlo模拟比较了各种点估计方法的精度.基于对数变换通过一阶泰勒展开,将两参数BS疲劳寿命分布BS(α,β)近似看作两参数对数正态分布,由此得到了2个参数 α,β的近似区间估计,通过M onte-Carlo模拟发现,所给出的近似方法比原有方法更精确.最后通过若干实例说明了方法的可行性.%The logarithmic moment estimations of parameters are proposed by logarithmic transformation for two-parameter Birnbaum-Saunders(BS) fatigue life distribution BS(α,β) under the full sample .The precisions of various point estimation methods are compared by a large number of Monte-Carlo simulations .Two-parameter BS fatigue life distribution BS(α,β) is approximately regarded as two-parameter lognormal distribution through the first order Taylor expansion based on logarithmic transformation . Then , the approximate interval estimations of two parameters α,βare obtained ,and it can be found that this approximate method is more accurate than the original method by Monte-Carlo simulations .Finally ,several examples show the feasibility of the methods .
展开▼
