设{Xn,n≥1}是一严平稳的ρ-混合的正的随机变量序列,且EX1=μ>0, Var(X1)=σ^2,记Sn=Σ(i=1)^n Xi和γ=σ/μ,在较弱的条件下,证明了对任意的x, limn→∞ 1/logn∑k=1^n 1/k I{(∏j=1^kSj/k!μ^k)^1/(γσ1√k≤k}=F(x),a.s., 其中σ1^2=1+2/(σ^2)∑(j=2)∞Cov(X1,Xj),F(·)是随机变量e^(2(1/2)N)的分布函数,N是标准正态随机变量,我们的结果推广了i.i.d时的情形.
Let {Xn,n≥1} be a strictly stationary p-mixing sequence of positive random X variables with EX1=μ〉0, and Var(X1)=σ^2,Denote Sn=Σ(i=1)^n Xi and γ=σ/μ, Under suitable conditions, we show that for any x, limn→∞ 1/logn∑k=1^n 1/k I{(∏j=1^kSj/k!μ^k)^1/(γσ1√k≤k}=F(x),a.s., where σ1^2=1+2/(σ^2)∑(j=2)∞Cov(X1,Xj),F(·) is the distribution function of the random variable e^(2(1/2)N) is a standard normal random variable. The result of Khurelbaatar and Rempata is a special case of ours.