【摘要】 定义Ｆ＝ｍｉｎ（ｔ＞ ０，Ｆ（ｔ）＝ １｝；设 ｔ＜ Ｆ为固定正实数， ｎ为样本总数， δ２Ｆ为Ｆ（ｎ）（ｔ）的方差，本文在假设Ｆ，Ｇ都连续可微的条件下得到如下结果：更多还原

【Abstract】 Let F = min{t> 0,F(t) =1}, t < F be fixed, F and G be continuous defferentiable functions, F2 be the variance of F(n)(t). We. show, in this paper, that the Edgeworth expansion of the KM estimator has the form P(δ－１Ｆ［[F(t)-F(n)(t)] ≤x) = Φ(x) + Ψn(x) + o(n-1), where φ(x) is the standard normal distribution function, φ is the standard normal density function, and更多还原