張永軍
(合肥學(xué)院 學(xué)報(bào)編輯部, 安徽 合肥 230601)
?
END隨機(jī)變量序列的完全收斂性
張永軍
(合肥學(xué)院 學(xué)報(bào)編輯部, 安徽 合肥 230601)
[摘要]利用END隨機(jī)變量的矩不等式, 研究END隨機(jī)變量序列的完全收斂性, 所得結(jié)果推廣了獨(dú)立隨機(jī)變量及若干相依變量的相應(yīng)結(jié)果。
[關(guān)鍵詞]END序列; 矩不等式; 完全收斂性
1預(yù)備知識(shí)
定義1對(duì)隨機(jī)變量序列{Xn,n≥1}, 若存在常數(shù)M>0, 對(duì)于任意的n≥1和一切實(shí)數(shù)x1,x2,…,xn都有
(1)
(2)
同時(shí)成立, 則稱{Xn,n≥1}為Extended Negatively Dependent(END)序列。
END序列的概念由文獻(xiàn)[2]給出,當(dāng)M=1,END序列即為NOD序列。而NOD隨機(jī)變量的概念又比NA隨機(jī)變量寬泛,文獻(xiàn)[3-7]給出NOD隨機(jī)變量序列的相關(guān)性質(zhì)和極限定理。END序列是一類比NA序列及NOD序列更廣泛的相依變量,研究其極限性質(zhì)和應(yīng)用具有較好的理論和實(shí)際意義。文獻(xiàn)[2]獲得了END重尾隨機(jī)變量的精確大偏差, 文獻(xiàn)[8]給出了END重尾隨機(jī)變量偏差的充分必要條件, 文獻(xiàn)[9]和[10]得到END隨機(jī)陣列加權(quán)和的完全收斂性結(jié)果。
本文中, 設(shè){Xn,n≥1}是定義在概率空間(Ω,F,P)上的END隨機(jī)變量序列。 C是正常數(shù), 在不同的地方可取不同的值。
引理 1[2]設(shè)隨機(jī)變量X1,X2,…,Xn為END隨機(jī)變量,若f1,f2,…,fn同為非降(或非增)函數(shù),則f1(X1),f2(X2),…,fn(Xn)仍為END隨機(jī)變量。
引理2[11]設(shè){Xn,n≥1}是均值為零的END序列,則存在僅依賴于p的正常數(shù)Cp,使得當(dāng)p≥2時(shí)有
(3)
2主要結(jié)果
定理1設(shè){Xn,n≥1}是均值為零的END序列,{an,n≥1}為正實(shí)數(shù)序列且an↑∞。設(shè){gn(t),n≥1}為取正值的偶函數(shù)序列。假定存在常數(shù)β∈(1,2]及δ>0,使得gn(t)≥δtβ,對(duì)0
(4)
則有
(5)
(6)
由(6)式又可推出
(7)
根據(jù)已知條件EXi>0,及對(duì)t>1,有g(shù)n(t)≥δt,可知
將上式結(jié)合(5)式,則有
(8)
利用(7)式和(8)式,知對(duì)充分大n,有
(9)
因此為證(5)式成立,只需證
(10)
(11)
由(4)式,即有
(10)式得證。
再由已知條件,存在1<β≤2,使得gn(t)≥δtβ,對(duì)0 可得 (11)式得證。 結(jié)合(9)—(11)式得定理1結(jié)論成立。 (12) 則有 注2.1END是一類更寬泛的隨機(jī)變量序列,它包括獨(dú)立序列,NA序列及NOD序列等。本文所得結(jié)論推廣了獨(dú)立隨機(jī)變量序列及NOD等相依序列的結(jié)論。 [參考文獻(xiàn)] [1]Hsu P L, Robbins H.Complete convergence and the law of large numbers[J]. Proceedings of the National Academy of Sciences of the United States of America,1947, 33(2): 25-31. [2]Liu L. Precise large deviations for dependent random variables with heavy tails[J]. Statistics and Probability Letters, 2009, 79(9): 1290-1298. [3]Joag-Dev K, Proschan F. Negative association of random variables with applications[J]. The Annals of Statistics, 1983, 11(1): 286-295. [4]Ghosh M. Multivariate negative dependence[J]. Communication in Statistics Theory and Methods, 1981, 10: 307-337. [5]Block H W, Savits T H, Shaked M. Some concepts of negative dependence[J]. The Annals of Statistics, 1982, 10(3): 765-772. [6]Alam K, Saxena K M L. Positive dependence in multivariate distributions[J]. Communication in Statistics-Theory and Methods, 1981, 10(12): 1183-1196. [7]Wu Y F, Zhu D J. Convergence properties of partial sums for arrays of rowwise negatively orthant dependent random variables[J]. Journal of the Korean Statistical Society, 2010, 39: 189-197. [8]Liu L. Necessary and sufficient conditions for moderate deviations of dependent random variables with heavy tails[J]. Science China Mathematics, 2010, 53(6): 1421-1434. [9]Qiu D H, Chen P Y, Antonini R G, Volodin A. On the complete convergence for arrays of rowwise extended negatively dependent random variables[J]. Journal of the Korean Mathematical Society, 2013, 50(2): 379-392. [10]Wu Y F, Guan M. Convergence properties of the partial sums for sequences of END random variables[J]. Journal of the Korean Mathematical Society, 2013, 49(6): 1097-1110. [11]Zhang G Z. Complete convergence for Sung's type weighted sums of END random variables[J]. Journal of Inequalities and Applications, 2014,2014: 353. Complete Convergence for Sequence of END Random Variable ZHANG Yongjun (EditorialDepartmentofJournal,HefeiUniversity,Hefei230601,China) Abstract:By using some moment inequality for END random variables, some complete convergence theorems for sequence of END random variable are obtained. These results generalize the corresponding theorems for independent sequence and some other type of dependent sequence. Key words:extended negatively dependent sequence; moment inequality; complete convergence [收稿日期]2016-02-05 [基金項(xiàng)目]2013年度合肥學(xué)院科研發(fā)展基金重點(diǎn)項(xiàng)目(13KY05ZD)資助。 [作者簡(jiǎn)介]張永軍(1975-),男,安徽合肥人,合肥學(xué)院學(xué)報(bào)編輯部副編審。 [中圖分類號(hào)]O211.4 [文獻(xiàn)標(biāo)識(shí)碼]A [文章編號(hào)]1674-2273(2016)03-0010-02