張文婷1,蔡際盼1,李永明
(1.廣西師范學(xué)院 數(shù)學(xué)科學(xué)學(xué)院, 廣西 南寧 530023;2.上饒師范學(xué)院,江西 上饒 334001)
引言
設(shè){Xn; n≥1)是概率空間(Ω;F;P)上的隨機(jī)變量序列, 具有相同的分布函數(shù)F(x)=P(X≤x).對(duì)于p∈(0,1),定義ζp=inf{x:F(x)≥p}為F(x)的p階分位數(shù), 記為F-1(p),其中函數(shù)F-1(t)(0 定義記Fu=σ(Xi,i∈u?N),N為自然數(shù)集.L2(Fu)表示所有Fu可測(cè)且二階矩有限的隨機(jī)變量全體,d(u,v)表示有限子集u和v的距離,令 下面我們給出一些基本假設(shè): (A2) F(x) 在ζp的某個(gè)鄰域Np內(nèi)可導(dǎo),密度函數(shù)滿足0 本文的主要結(jié)果如下: 定理1 假設(shè)(A1),(A2)成立,log n 表示以2為底的對(duì)數(shù),當(dāng)n→∞時(shí), 有 定理2 滿足引理5和定理1的條件,當(dāng)n→∞時(shí), 有 下面我們給出本文所需引理. 引理2[11]設(shè)F(x)是右連續(xù)的分布函數(shù),則廣義逆函數(shù)F-1(t), 在0 (1) F-1(F(x))≤x,-∞ 引理3[12]令p∈(0,1),ζp,n=Fn-1(p)=inf{x:Fn(x)≥p},假設(shè)P(Xi=Xj)=0,i≠j. 那么 證明由于Fn(x) 是非降函數(shù), 可得 =D1+D2. (3.1) (3.2) 根據(jù)微分中值定理, 有 (3.3) 因此, 由(3.1)-(3.3) 可得 引理得證. 引理6 滿足引理5的條件, 當(dāng)n→∞時(shí), 有 證明根據(jù)引理5可得 引理證畢. 定理1的證明令k≥1, 根據(jù)子序列法要證結(jié)論成立只需證明 (4.1) 下面我們來證明(4.1)式. 由于 =H1+H2. (4.2) 根據(jù)引理2,引理3,引理4和Markov不等式以及Taylor 公式, 有 (4.3) (4.4) 因此, 由(4.2)(4.3)(4.4)式可得 故根據(jù)Borel-Cantelli 引理可得(4.1) 式成立, 從而該定理得證. 定理2的證明根據(jù)引理3, 有 Fn(ζp,n)-p=O(n-1),a.s.. (4.5) 由引理6可得 (4.6) 從而, 根據(jù)(4.5)(4.6) 式和定理1以及Taylor 公式可得 其中ηn是介于ζp,n與ζp之間的隨機(jī)變量. 整理上式可得 定理證明完畢. 參考文獻(xiàn): [1] Shuxia Sun. The Bahadur representation for sample quantiles under weak dependence [J]. Statistics and Probability Letters, 2006, 76: 1238-1244. [2] Ajami M., Fakoor V., Jomhoori S. The Bahadur representation for kernel-type estimator of the quantile function under strong mixing and censored data [J]. Statistics and Probability Letters,2011, 81(8): 1306-1310. [3] Guodong Xing, Shanchao Yang, Yan Liu, Keming Yu. A note on the Bahadur representation of sample quantiles for α-mixing random variables [J]. Monatshefte für Mathematik, 2012, 165:579-596. [4] Qinchi Zhang, Wenzhi Yang, Shuhe hu. On Bahadur representation for sample quantiles under α-mixing sequence [J]. Statistical Papers, 2012. [5] Guodong Xing, Shanchao Yang. A remark on the Bahadur representation of sample quantiles for negatively associated sequences [J]. Journal of the Korean Statistical Society, 2011, 40(3): 277-280. [6] 梁丹, 楊善朝, 蒙玉波. NOD 序列樣本分位數(shù)的Bahadur 表示[J]. 工程數(shù)學(xué)學(xué)報(bào), 2013, 30(1): 77-84. [7] 吳群英. 混合序列的概率極限理論[M]. 北京: 科學(xué)出版社, 2006. [9] Yongfeng Wu, Chunhua Wang, Andrei Volodin. Limiting behavior for arrays of rowwise ρ*Mixing random variables[J]. Lithuanian Mathematical Journal, 2012, 52(2): 214-221. [10] Mingle Guo, Dongjin Zhu.Complete convergence of weighted sums for ρ*mixing sequence of random variables[J]. Journal of Mathematical Research with Applications, 2013, 33(4): 483-492. [11] Serfling R.J. Approximation theorems of mathematical statistics [M]. New York: John Wiley and Sons, 1980. [12] Xuejun Wang, Shuhe Hu, Wenzhi Yang. The Bahadur representation for sample quantiles under strongly mixing sequence [J]. Journal of Statistical Planning and Inference, 2011, 141: 655-662. [13] Sergey Utev, Magda Peligrad. Maximal inequalities and an invariance principle for a class of weakly dependent random variables[J]. Journal of Theoretical Probability, 2003, 16(1): 101-115.1 基本假設(shè)和結(jié)論
2 輔助結(jié)論
3 主要結(jié)論的證明