Skip to main content

A note on complete convergence of weighted sums for array of rowwise AANA random variables

Abstract

In this paper, we consider complete convergence and complete moment convergence of weighted sums for an array of rowwise AANA random variables. The main result of the paper generalizes the Baum-Katz theorem on AANA random variables. Our results extend and improve the corresponding ones of Wang et al. (Abstr. Appl. Anal. 2012:315138, 2012).

MSC:60B10, 60F15.

1 Introduction

Assume that random variables X n , nN={1,2,} are defined on a fixed probability space (Ω,A,P).

First, we recall two definitions as follows.

Definition 1.1 Random variables X 1 , X 2 ,, X n , n2, are said to be negatively associated (NA, in short) if

Cov ( f ( X i 1 , , X i k ) , g ( X j 1 , , X j m ) ) 0

for any pair of nonempty disjoint subsets A={ i 1 ,, i k } and B={ j 1 ,, j m }, k+mn, of the set {1,2,,n} and for any bounded coordinatewise increasing real functions f( x i 1 ,, x i k ) and g( x j 1 ,, x j m ), x 1 ,, x n R=(,). Random variables X n , nN, are NA if every nN random variables X 1 , X 2 ,, X n are NA.

Random variables X n i , i,nN, are called an array of rowwise NA random variables if for every nN random variables X n i , iN are, NA.

The concept of NA random variables was introduced by Block et al. [1] and carefully studied by Joav-Dev and Proschan [2]. Primarily motivated by this, Chandra and Ghosal [3, 4] introduced the following dependence.

Definition 1.2 Random variables X n , nN, are said to be asymptotically almost negatively associated (AANA, in short) if there exists a nonnegative sequence q(n)0 as n such that

Cov ( f ( X n ) , g ( X n + 1 , , X n + k ) ) q(n) { Var ( f ( X n ) ) Var ( g ( X n + 1 , , X n + k ) ) } 1 / 2

for all n,kN and for all coordinatewise nondecreasing continuous functions f and g for which Varf( X n ) and Varg( X n + 1 ,, X n + k ) exist.

Random variables X n i , i,nN, are called an array of rowwise AANA random variables if for every nN, random variables X n i , iN, are AANA.

The family of AANA random variables contains NA (in particular, independent) random variables (with q(n)=0, n1) and some more kinds of random variables which are not much deviated from being negatively associated. An example of AANA random variables which are not NA was constructed by Chandra and Ghosal [3]. For various results and applications of AANA random variables, one can refer to Chandra and Ghosal [4], Wang et al. [5], Ko et al. [6], Yuan and An [7], Wang et al. [8, 9] and Wang et al. [10], Yang et al. [11], Shen and Wu [12] among others.

The concept of complete convergence was introduced by Hsu and Robbins [13] as follows. Random variables U n , nN, are said to converge completely to a constant C if n = 1 P(| U n C|>ε)< for all ε>0. In view of the Borel-Cantelli lemma, this implies that U n C almost surely (a.s.). The converse is true if random variables U n , nN, are independent. Hsu and Robbins [13] proved that arithmetic means of independent and identically distributed (i.i.d.) random variables converges completely to the expected value if the variance of the summands is finite. Erdös [14] proved the converse. The result of Hsu, Robbins and Erdös is a fundamental theorem in probability theory and has been generalized and extended in several directions by many authors. One of the most important generalizations was provided by Baum and Katz [15] for the strong law of large numbers as follows.

Theorem A Let 1/2<α1 and αp>1. Let X n , nN be i.i.d. random variables with zero means. Then the following statements are equivalent:

  1. (i)

    E | X 1 | p <,

  2. (ii)

    n = 1 n α p 2 P( max 1 j n | i = 1 j X i |>ε n α )< for all ε>0.

Motivated by Baum and Katz [15] for i.i.d. random variables, many authors studied the Baum-Katz-type theorem for dependent random variables. One can refer to Peligrad [16], Shao [17], Peligrad and Gut [18], Kruglov et al. [19], Wang and Hu [20], Shen et al. [21], Wang et al. [22], etc.

Next, we will give the definition of stochastic domination which is used frequently in the paper.

Definition 1.3 Random variables X n , nN, are said to be stochastically dominated by a random variable X if for every nN there exists a positive constant C such that

P ( | X n | > x ) CP ( | X | > x )

for all x0.

An array of rowwise random variables X n i , i,nN, is said to be stochastically dominated by a random variable X if for every nN there exists a positive constant C such that

sup i 1 P ( | X n i | > x ) CP ( | X | > x )

for all x0.

Wang et al. [10] discussed the complete convergence for an array of rowwise AANA random variables which are stochastically dominated by a random variable X and obtained the following result.

Theorem B Let X n i , i,nN, be an array of rowwise AANA random variables which are stochastically dominated by a random variable X and E X n i =0 for every i,nN with q(n) from Definition  1.2.

  1. (i)

    Let 1/2<α1, p>1 and αp>1. If E | X | p < and n = 1 q s / r (n)< for some r(3 2 k 1 ,4 2 k 1 ] and

    r>max ( 2 , α p 1 α 1 / 2 , p ) ,

where integer number k1 and sr/(r1) for r>1, then for all ε>0,

n = 1 n α p 2 P ( max 1 j n | i = 1 j X n i | > ε n α ) <.
  1. (ii)

    If E|X|log|X|< and n = 1 q 2 (n)<, then for all ε>0,

    n = 1 n 1 P ( max 1 j n | i = 1 j X n i | > ε n ) <.

The complete convergence for an array of rowwise random variables was studied by many authors. See, for example, the complete convergence for an array of rowwise independent random variables was studied by Hu et al. [23], Sung et al. [24], Kruglov et al. [19] and others. Recently, many authors extended the complete convergence for an array of rowwise independent random variables to the cases of dependent random variables. One can refer to Kuczmaszewska [25, 26], Chen et al. [27], Kruglov [28], Zhou and Lin [29], Guo [30], Wu [31], and so on.

The main purpose of the paper is to further study the complete convergence and complete moment convergence of weighted sums for an array of rowwise AANA random variables. The result of the paper generalizes the Baum-Katz theorem on AANA random variables in different methods. As an application, we get the Marcinkiewicz-Zygmund type strong law of large numbers for weighted sums on AANA random variables. Our results extend and improve the corresponding ones of [10].

Throughout this paper, for r>1, let sr/(r1) be the dual number of r. The symbols C, C 1 , C 2 , denote positive constants which may be different in various places. Assume that I(A) is the indicator function of the set A. Let x + =max(0,x) and logx=lnmax(x,e), where lnx denotes the natural logarithm. a n =O( b n ) stands for | a n |C| b n |.

2 Preliminaries

To prove the main results of the paper, we need the following lemmas.

Lemma 2.1 (cf. [[32], Lemma 4.1.6])

Let X n , nN, be random variables, which are stochastically dominated by a random variable X. Then, for any a>0 and b>0, the following two statements hold:

E | X n | a I ( | X n | b ) C 1 { E | X | a I ( | X | b ) + b a P ( | X | > b ) }

and

E | X n | a I ( | X n | > b ) C 2 E | X | a I ( | X | > b ) ,

where C 1 and C 2 are positive constants.

Lemma 2.2 (cf. [[7], Lemma 2.1])

Let X n , nN, be AANA random variables with q(n) from Definition  1.2. Assume that f n , nN are all nondecreasing (or all nonincreasing) and continuous functions, then f n ( X n ), nN, are still AANA random variables with q(n).

Lemma 2.3 (cf. [[7], Theorem 2.1])

Let r>1 and X n , nN, be AANA random variables with q(n) from Definition  1.2.

If n = 1 q 2 (n)<, then there exists a positive constant C r depending only on r such that for all n1 and 1<r2,

E ( max 1 j n | i = 1 j X i | r ) C r i = 1 n E | X i | r .

If n = 1 q s / r (n)< for some r(3 2 k 1 ,4 2 k 1 ], where integer number k1, then there exists a positive constant D r depending only on r such that for all n1,

E ( max 1 j n | i = 1 j X i | r ) D r { i = 1 n E | X i | r + ( i = 1 n E X i 2 ) r / 2 } .

Lemma 2.4 (cf. [[33], Lemma 2.4])

Let Y n , Z n , nN be random variables. Then, for any q>1, ε>0 and a>0,

E ( max 1 j n | i = 1 j ( Y i + Z i ) | ε a ) + ( 1 ε q + 1 q 1 ) 1 a q 1 E max 1 j n | i = 1 j Y i | q +E max 1 j n | i = 1 j Z i | .

3 Main results and their proofs

In this section, let X n i , i,nN, be an array of rowwise AANA random variables, i.e., for every nN, X n i , iN, are AANA random variables with the identical mixing coefficient q(i) and let a n i , i,nN, be an array of real numbers. Let X n , nN, be AANA random variables with q(n) from Definition 1.2.

In the following, let ψ(x)=1 or ψ(x)=logx. Note that the function ψ(x) has the following properties (see [34]):

  1. (a)

    for all mk1,

    n = k m n r 1 ψ(n)C m r ψ(m)if r>0
    (3.1)

and

n = m n r 1 ψ(n)C m r ψ(m)if r<0;
(3.2)
  1. (b)

    for all p>0,

    ψ ( | x | p ) C(p)ψ ( | x | ) C(p)ψ ( 1 + | x | ) .
    (3.3)

We will consider the following conditions.

(H1) n = 1 q s / r (n)< for some r(3 2 k 1 ,4 2 k 1 ] and r> α p 1 α 1 / 2 , where integer number k1 if α>1/2, αp>1 and p2.

(H2) n = 1 q 2 (n)< if α>1/2, αp>1 and 1p<2 or α>1/2 and αp=1.

Theorem 3.1 Let α> 1 2 and αp1. Assume that X n i , i,nN, are an array of rowwise AANA random variables which are stochastically dominated by a random variable X, a n i , i,nN, are an array of real numbers with i = 1 n | a n i | q =O(n) for some q>max{ α p 1 α 1 / 2 ,2}. Let E X n i =0 for all i,nN, if p1 and the conditions (H1) and (H2) are satisfied. If

E | X | p ψ ( | X | ) <,
(3.4)

then

n = 1 n α p 2 ψ(n)P ( max 1 j n | i = 1 j a n i X n i | > ε n α ) < for all ε>0.
(3.5)

Proof Without loss of generality, we can assume that a n i >0 for all i,nN. For fixed nN, let X n i = n α I( X n i < n α )+ X n i I(| X n i | n α )+ n α I( X n i > n α ) and X n i = X n i X n i , i1. We will consider the following three cases.

  1. (i)

    Let p>1. It is easy to check that

    n = 1 n α p 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i X n i | > ε n α ) n = 1 n α p 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i ( X n i E X n i ) | > ε n α / 2 ) + n = 1 n α p 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i ( X n i E X n i ) | > ε n α / 2 ) : = I + J .

By C r inequality and i = 1 n a n i q =O(n), it is easy to check that for all 0<γq,

1 n i = 1 n a n i γ ( 1 n i = 1 n a n i q ) γ / q =O(1).
(3.6)

For J , noting that | X n i || X n i |I(| X n i |> n α ), we have by Markov’s inequality, Lemma 2.1 and (3.3), that

J C n = 1 n α p 2 α ψ ( n ) i = 1 n a n i E | X n i | C n = 1 n α p 2 α ψ ( n ) i = 1 n a n i E | X n i | I ( | X n i | > n α ) C n = 1 n α p 1 α ψ ( n ) E | X | I ( | X | > n α ) = C n = 1 n α p 1 α ψ ( n ) j = n E | X | I ( j < | X | 1 / α j + 1 ) = C j = 1 E | X | I ( j < | X | 1 / α j + 1 ) n = 1 j n α p 1 α ψ ( n ) C j = 1 j α p α ψ ( j ) E | X | I ( j < | X | 1 / α j + 1 ) C E | X | p ψ ( | X | 1 / α ) C E | X | p ψ ( | X | ) < .
(3.7)

For I , note that for every nN, a n i X n i E a n i X n i , iN, are AANA random variables from Lemma 2.2. By Markov’s inequality, Lemma 2.3 and Jensen’s inequality, we have that for any r2,

I C r n = 1 n α p 2 α r ψ ( n ) E ( max 1 j n | i = 1 j ( a n i X n i E a n i X n i ) | r ) C r n = 1 n α p 2 α r ψ ( n ) i = 1 n a n i r E | X n i | r + C r n = 1 n α p 2 α r ψ ( n ) ( i = 1 n a n i 2 E ( X n i ) 2 ) r / 2 : = I 1 + I 2 .
(3.8)

We consider the following three cases.

Case 1. α>1/2, αp>1 and p2.

Take r=q. By q>max{ α p 1 α 1 / 2 ,2}, it follows that q>p and αp2αq+q/2<1.

For I 1 , we have by C r inequality that

I 1 C n = 1 n α p 2 α q ψ ( n ) i = 1 n a n i q ( E | X n i | q I ( | X n i | n α ) + n α q P ( | X n i | > n α ) ) C n = 1 n α p 2 α q ψ ( n ) i = 1 n a n i q ( E | X | q I ( | X | n α ) + n α q P ( | X | > n α ) ) C n = 1 n α p 1 α q ψ ( n ) E | X | q I ( | X | n α ) + C n = 1 n α p 1 α ψ ( n ) E | X | I ( | X | > n α ) C n = 1 n α ( p q ) 1 ψ ( n ) j = 1 n j α q P ( j 1 < | X | 1 / α j ) + C E | X | p ψ ( | X | ) C j = 1 j α q P ( j 1 < | X | 1 / α j ) n = j n α ( p q ) 1 ψ ( n ) + C E | X | p ψ ( | X | ) C j = 1 j α p ψ ( j ) P ( j 1 < | X | 1 / α j ) + C E | X | p ψ ( | X | ) C E | X | p ψ ( | X | 1 / α ) C E | X | p ψ ( | X | ) < .
(3.9)

For I 2 , note that E X 2 < if E | X | p ψ(|X|)< for p2. We have by (3.6) that

I 2 C n = 1 n α p 2 α q ψ ( n ) ( i = 1 n a n i 2 E X n i 2 ) q / 2 C n = 1 n α p 2 α q ψ ( n ) ( i = 1 n a n i 2 E X 2 ) q / 2 C n = 1 n α p 2 α q + q / 2 ψ ( n ) < .

Case 2. α>1/2, αp>1 and 1<p<2.

Take r=2. Similar to the proofs of (3.8), (3.9) and (3.7), we have that

I C n = 1 n α p 2 2 α ψ ( n ) i = 1 n a n i 2 ( E X n i 2 I ( | X n i | n α ) + n 2 α P ( | X n i | > n α ) ) C n = 1 n α p 1 2 α ψ ( n ) E X 2 I ( | X | n α ) + C n = 1 n α p 1 α ψ ( n ) E | X | I ( | X | > n α ) < .
(3.10)

Case 3. α>1/2, αp=1 and p>1.

Take r=2. Note that 1/2<α<1 if αp=1. Similar to the proof of (3.10), it follows that I <.

  1. (ii)

    Let p=1. Note that α1 from αp1. By E X n i =0 for i,nN, Lemma 2.1, (3.6) and (3.4), we have that

    n α max 1 j n | i = 1 j a n i E X n i | n α i = 1 n a n i E | X n i | I ( | X n i | > n α ) n 1 α E | X | I ( | X | > n α ) 0 as  n .

Hence for n large enough, we have

n α max 1 j n | i = 1 j a n i X n i | < ε 2 .
(3.11)

It follows that

n = 1 n α 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i X n i | > ε n α ) n = 1 n α 2 ψ ( n ) i = 1 n P ( | X n i | > n α ) + n = 1 n α 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i X n i | > ε n α ) C n = 1 n α 1 ψ ( n ) P ( | X | > n α ) + C n = 1 n α 2 ψ ( n ) P ( max 1 j n | i = 1 j a n i ( X n i E X n i ) | > ε n α 2 ) : = C I 1 + C I 2 .
(3.12)

For I 1 , we have by (3.1) and (3.4) that

I 1 = n = 1 n α 1 ψ ( n ) i = n P ( i α < | X | ( i + 1 ) α ) = i = 1 P ( i α < | X | ( i + 1 ) α ) n = 1 i n α 1 ψ ( n ) C i = 1 P ( i α < | X | ( i + 1 ) α ) i α ψ ( i ) C E | X | ψ ( | X | 1 / α ) C E | X | ψ ( | X | ) < .
(3.13)

For I 2 , we have by Markov’s inequality, Lemma 2.3, Lemma 2.1, (3.2) and (3.3) that

I 2 C n = 1 n α 2 ψ ( n ) E max 1 j n ( i = 1 j a n i ( X n i E X n i ) ) 2 C n = 1 n α 2 ψ ( n ) i = 1 n a n i 2 E ( X n i ) 2 = C n = 1 n α 2 ψ ( n ) { i = 1 n a n i 2 E X n i 2 I ( | X n i | n α ) + n 2 α i = 1 n a n i 2 P ( | X n i | > n α ) } C n = 1 n α 1 ψ ( n ) E X 2 I ( | X | n α ) + C n = 1 n α 1 ψ ( n ) P ( | X | > n α ) = C n = 1 n α 1 ψ ( n ) k = 1 n E X 2 I ( ( k 1 ) α < | X | k α ) + C = C k = 1 E X 2 I ( ( k 1 ) α < | X | k α ) n = k n α 1 ψ ( n ) + C C k = 1 k α ψ ( k ) E X 2 I ( ( k 1 ) α < | X | k α ) + C C E | X | ψ ( | X | ) + C < .
(3.14)

By (3.12)-(3.14), (3.5) holds for the case p=1.

  1. (iii)

    Let 0<p<1. Denote

    i = 1 j a n i X n i = i = 1 j a n i X n i I ( | X n i | n α ) + i = 1 j a n i X n i I ( | X n i | > n α ) =: S n j + S n j .
    (3.15)

Noting that E | X | p ψ(|X|)<, we have by Markov’s inequality, Lemma 2.1 and (3.2)-(3.6), that

n = 1 n α p 2 ψ ( n ) P ( max 1 j n | S n j | > ε n α ) ε 1 n = 1 n α p 2 α ψ ( n ) E ( max 1 j n | i = 1 j a n i X n i I ( | X n i | n α ) | ) ε 1 n = 1 n α p 2 α ψ ( n ) i = 1 n a n i E | X n i | I ( | X n i | n α ) C ε 1 n = 1 n α p 1 α ψ ( n ) E | X | I ( | X | n α ) + C ε 1 n = 1 n α p 1 ψ ( n ) P ( | X | > n α ) = C ε 1 n = 1 n α p 1 α ψ ( n ) j = 1 n E | X | I ( j 1 < | X | 1 / α j ) + C ε 1 n = 1 n α p 1 ψ ( n ) j = n P ( j < | X | 1 / α j + 1 ) C ε 1 j = 1 j α P ( j 1 < | X | 1 / α j ) n = j n α p 1 α ψ ( n ) + C ε 1 j = 1 P ( j < | X | 1 / α j + 1 ) n = 1 j n α p 1 ψ ( n ) C ε 1 j = 1 j α p ψ ( j ) P ( j 1 < | X | 1 / α j ) + C ε 1 j = 1 j α p ψ ( j ) P ( j < | X | 1 / α j + 1 ) C E | X | p ψ ( | X | 1 / α ) C E | X | p ψ ( | X | ) <
(3.16)

and

n = 1 n α p 2 ψ ( n ) P ( max 1 j n | S n j | > ε n α ) ε p / 2 n = 1 n α p / 2 2 ψ ( n ) E ( max 1 j n | i = 1 j a n i X n i I ( | X n i | > n α ) | ) p / 2 ε p / 2 n = 1 n α p / 2 2 ψ ( n ) i = 1 n a n i p / 2 E | X n i | p / 2 I ( | X n i | > n α ) C ε p / 2 n = 1 n α p / 2 1 ψ ( n ) E | X | p / 2 I ( | X | > n α ) = C ε p / 2 n = 1 n α p / 2 1 ψ ( n ) j = n E | X | p / 2 I ( j < | X | 1 / α j + 1 ) C ε p / 2 j = 1 j α p / 2 P ( j < | X | 1 / α j + 1 ) n = 1 j n α p / 2 1 ψ ( n ) C ε p / 2 j = 1 j α p ψ ( j ) P ( j 1 < | X | 1 / α j ) C E | X | p ψ ( | X | 1 / α ) C E | X | p ψ ( | X | ) < .
(3.17)

Hence (3.15)-(3.17) imply (3.5). From all the statements above, we have proved (3.5). □

Remark 3.1 Taking ψ(x)1 and a n i 1 in Theorem 3.1, we can get (i) of Theorem B; meanwhile, relax the mixing coefficient condition n = 1 q s / r (n)< to n = 1 q 2 (n)< for the case α>1/2, αp>1 and 1<p<2. In addition, we extend the case 1/2<α1, p>1 and αp>1 to the case α>1/2, αp1. Taking ψ(x)1, a n i 1 and α=1, p=1 in Theorem 3.1, we can get (ii) of Theorem B and weaken the condition E|X|log|X|< to the condition E|X|<. Hence we extend and improve the corresponding results of [10].

Remark 3.2 Under the conditions of Theorem 3.1, we have that for p>1,

n = 1 n α p 2 α ψ(n)E ( max 1 j n | i = 1 j a n i X n i | ε n α ) + <.
(3.18)

In fact, by Lemma 2.4 with r2, we get

n = 1 n α p 2 α ψ ( n ) E ( max 1 j n | i = 1 j a n i X n i | ε n α ) + C n = 1 n α p 2 α r ψ ( n ) E ( max 1 j n | i = 1 j ( a n i X n i E a n i X n i ) | ) r + n = 1 n α p 2 α ψ ( n ) E ( max 1 j n | i = 1 j ( a n i X n i E a n i X n i ) | ) .

By the process of the proof of Theorem 3.1 in the case p>1, it follows that (3.18) holds.

Similar to the proof of Theorem 3.1, we can get easily the following result.

Theorem 3.2 Let α> 1 2 and αp1. Let X n , nN, be AANA random variables which are stochastically dominated by a random variable X. Assume that a n , nN, are real numbers with i = 1 n | a i | q =O(n) for some q>max{ α p 1 α 1 / 2 ,2}, the conditions (H1) and (H2) are satisfied. If (3.4) holds, then

n = 1 n α p 2 ψ(n)P ( max 1 j n | i = 1 j a i X i | > ε n α ) < for all ε>0.

Remark 3.3 Similar to Remark 3.1, taking ψ(x)1 and a i 1 in Theorem 3.2, we can get (i) of Theorem 3.4 in [10]; meanwhile, relax the mixing coefficient condition n = 1 q s / r (n)< to n = 1 q 2 (n)< for the case α>1/2, αp>1 and 1<p<2. In addition, we extend the case 1/2<α1, p>1 and αp>1 to the case α>1/2, αp1. Taking ψ(x)1, a n i 1 and α=1, p=1 in Theorem 3.2, we can get (ii) of Theorem 3.4 in [10] and weaken the condition E|X|log|X|< to the condition E|X|<. Hence, we extend and improve the corresponding results of [10].

In the following, we give the Marcinkiewicz-Zygmund type strong law of large numbers of weights sums on AANA random variables.

Corollary 3.1 Let α> 1 2 and αp1. Let X n , nN, be AANA random variables which are stochastically dominated by a random variable X. Assume that a n , nN are real numbers with i = 1 n | a i | q =O(n) for some q>max{ α p 1 α 1 / 2 ,2}, the conditions (H1) and (H2) are satisfied. If E | X | p <, then

n = 1 n α p 2 P ( max 1 j n | i = 1 j a i X i | > ε n α ) <
(3.19)

and

n α i = 1 n a i X i 0 a.s. n.
(3.20)

Further, for p>1,

n = 1 n α p 2 α E ( max 1 j n | i = 1 j a i X i | ε n α ) + <.
(3.21)

Proof Taking ψ(x)=1 in Theorem 3.2, we get (3.19) easily. Similar to the proof of (3.18), (3.21) is obtained immediately. We only need to prove (3.20).

By (3.20), it follows that for all ε>0,

> n = 1 n α p 2 P ( max 1 j n | i = 1 j a i X i | > ε n α ) = k = 0 n = 2 k 2 k + 1 1 n α p 2 P ( max 1 j n | i = 1 j a i X i | > ε n α ) { k = 0 ( 2 k ) α p 2 2 k P ( max 1 j 2 k | i = 1 j a i X i | > ε 2 ( k + 1 ) α ) if  α p 2 , k = 0 ( 2 k + 1 ) α p 2 2 k P ( max 1 j 2 k | i = 1 j a i X i | > ε 2 ( k + 1 ) α ) if  1 α p < 2 { k = 0 P ( max 1 j 2 k | i = 1 j a i X i | > ε 2 ( k + 1 ) α ) if  α p 2 , 1 2 k = 0 P ( max 1 j 2 k | i = 1 j a i X i | > ε 2 ( k + 1 ) α ) if  1 α p < 2 .

By the Borel-Cantelli lemma, we obtain that

max 1 j 2 k | i = 1 j a i X i | 2 ( k + 1 ) α 0a.s. k.
(3.22)

For all positive integers n, there exists a positive integer k such that 2 k 1 n 2 k . We have by (3.22) that

n α | i = 1 n a i X i | max 2 k 1 n 2 k n α | i = 1 n a i X i | 2 α max 1 j 2 k | i = 1 j a i X i | 2 ( k + 1 ) α 0a.s. k,

which implies that

n α i = 1 n a i X i 0a.s. n.

This completes the proof of the corollary. □

Remark 3.4 Taking a n 1 in Corollary 3.1, we can get the Baum-Katz result on AANA random variables. Comparing with Theorem 3.4 and Corollary 3.5 of [10], Corollary 3.1 relaxes the mixing coefficient condition n = 1 q s / r (n)< to n = 1 q 2 (n)< for the case α>1/2, αp>1 and 1<p<2. In addition, we also consider the case αp=1 and the case αp1 and 0<p1. Taking α=1 and p=2 in Corollary 3.1, we can get the Hsu-Robbins-type theorem (see [13]) on AANA random variables. Taking α=1 and p=1 in Corollary 3.1, we improve (ii) of Theorem 3.4 and (ii) of Corollary 3.5 in [10].

References

  1. Block HW, Savits TH, Shaked M: Some concepts of negative dependence. Ann. Probab. 1982, 10(3):765–772. 10.1214/aop/1176993784

    Article  MathSciNet  Google Scholar 

  2. Joag-Dev K, Proschan F: Negative association of random variables, with applications. Ann. Stat. 1983, 11(1):286–295. 10.1214/aos/1176346079

    Article  MathSciNet  Google Scholar 

  3. Chandra TK, Ghosal S: Extensions of the strong law of large numbers of Marcinkiewicz and Zygmund for dependent variables. Acta Math. Hung. 1996, 71(4):327–336. 10.1007/BF00114421

    Article  MathSciNet  Google Scholar 

  4. Chandra TK, Ghosal S: The strong law of large numbers for weighted averages under dependence assumptions. J. Theor. Probab. 1996, 9(3):797–809. 10.1007/BF02214087

    Article  MathSciNet  Google Scholar 

  5. Wang Y, Yan J, Cheng F, Su C: The strong law of large numbers and the law of iterated logarithm for product sums of NA and AANA random variables. Southeast Asian Bull. Math. 2003, 27(2):369–384.

    MathSciNet  Google Scholar 

  6. Ko M-H, Kim T-S, Lin Z: The Hàjeck-Rènyi inequality for the AANA random variables and its applications. Taiwan. J. Math. 2005, 9(1):111–122.

    MathSciNet  Google Scholar 

  7. Yuan DM, An J: Rosenthal type inequalities for asymptotically almost negatively associated random variables and applications. Sci. China Ser. A 2009, 52(9):1887–1904. 10.1007/s11425-009-0154-z

    Article  MathSciNet  Google Scholar 

  8. Wang XJ, Hu SH, Yang WZ: Convergence properties for asymptotically almost negatively associated sequence. Discrete Dyn. Nat. Soc. 2010., 2010: Article ID 218380

    Google Scholar 

  9. Wang XJ, Hu SH, Yang WZ: Complete convergence for arrays of rowwise asymptotically almost negatively associated random variables. Discrete Dyn. Nat. Soc. 2011., 2011: Article ID 717126

    Google Scholar 

  10. Wang XJ, Hu SH, Yang WZ, Wang XH: On complete convergence of weighted sums for arrays of rowwise asymptotically almost negatively associated random variables. Abstr. Appl. Anal. 2012., 2012: Article ID 315138

    Google Scholar 

  11. Yang WZ, Wang XJ, Ling NX, Hu SH: On complete convergence of moving average process for AANA sequence. Discrete Dyn. Nat. Soc. 2012., 2012: Article ID 863931

    Google Scholar 

  12. Shen AT, Wu RC: Strong and weak convergence for asymptotically almost negatively associated random variables. Discrete Dyn. Nat. Soc. 2013., 2013: Article ID 235012

    Google Scholar 

  13. Hsu PL, Robbins H: Complete convergence and the law of large numbers. Proc. Natl. Acad. Sci. USA 1947, 33(2):25–31. 10.1073/pnas.33.2.25

    Article  MathSciNet  Google Scholar 

  14. Erdös P: On a theorem of Hsu and Robbins. Ann. Math. Stat. 1949, 20(2):286–291. 10.1214/aoms/1177730037

    Article  Google Scholar 

  15. Baum LE, Katz M: Convergence rates in the law of large numbers. Trans. Am. Math. Soc. 1965, 120(1):108–123. 10.1090/S0002-9947-1965-0198524-1

    Article  MathSciNet  Google Scholar 

  16. Peligrad M: Convergence rates of the strong law for stationary mixing sequences. Z. Wahrscheinlichkeitstheor. Verw. Geb. 1985, 70(2):307–314. 10.1007/BF02451434

    Article  MathSciNet  Google Scholar 

  17. Shao QM: A moment inequality and its applications. Acta Math. Sin. 1988, 31(6):736–747.

    Google Scholar 

  18. Peligrad M, Gut A: Almost-sure results for a class of dependent random variables. J. Theor. Probab. 1999, 12(1):87–104. 10.1023/A:1021744626773

    Article  MathSciNet  Google Scholar 

  19. Kruglov VM, Volodin A, Hu T-C: On complete convergence for arrays. Stat. Probab. Lett. 2006, 76: 1631–1640. 10.1016/j.spl.2006.04.006

    Article  MathSciNet  Google Scholar 

  20. Wang XJ, Hu SH: Some Baum-Katz type results for φ -mixing random variables with different distributions. Rev. R. Acad. Cienc. Exactas Fís. Nat., Ser. a Mat. (Ed. Impr.) 2012, 106(2):321–331.

    Article  Google Scholar 

  21. Shen AT, Wang XH, Ling JM: On complete convergence for non-stationary φ -mixing random variables. Commun. Stat., Theory Methods 2012. 10.1080/03610926.2012.725501

    Google Scholar 

  22. Wang XJ, Hu SH, Yang WZ, Wang XH: Convergence rates in the strong law of large numbers for martingale difference sequences. Abstr. Appl. Anal. 2012., 2012: Article ID 572493

    Google Scholar 

  23. Hu T-C, Ordóñez Cabrera M, Sung SH, Volodin A: Complete convergence for arrays of rowwise independent random variables. Commun. Korean Math. Soc. 2003, 18: 375–383.

    Article  MathSciNet  Google Scholar 

  24. Sung SH, Volodin A, Hu T-C: More on complete convergence for arrays. Stat. Probab. Lett. 2005, 71: 303–311. 10.1016/j.spl.2004.11.006

    Article  MathSciNet  Google Scholar 

  25. Kuczmaszewska A: On complete convergence for arrays of rowwise dependent random variables. Stat. Probab. Lett. 2007, 77: 1050–1060. 10.1016/j.spl.2006.12.007

    Article  MathSciNet  Google Scholar 

  26. Kuczmaszewska A: On complete convergence for arrays of rowwise negatively associated random variables. Stat. Probab. Lett. 2009, 79: 116–124. 10.1016/j.spl.2008.07.030

    Article  MathSciNet  Google Scholar 

  27. Chen P, Hu T-C, Liu X, Volodin A: On complete convergence for arrays of row-wise negatively associated random variables. Theory Probab. Appl. 2008, 52: 323–328. 10.1137/S0040585X97983079

    Article  MathSciNet  Google Scholar 

  28. Kruglov VM: Complete convergence for maximal sums of negatively associated random variables. J. Probab. Stat. 2010., 2010: Article ID 764043

    Google Scholar 

  29. Zhou XC, Lin JG: On complete convergence for arrays of rowwise ρ -mixing random variables and its applications. J. Inequal. Appl. 2010., 2010: Article ID 769201

    Google Scholar 

  30. Guo ML: Complete moment convergence of weighted sums for arrays of rowwise φ -mixing random variables. Int. J. Math. Math. Sci. 2012., 2012: Article ID 730962

    Google Scholar 

  31. Wu QY: A complete convergence theorem for weighted sums of arrays of rowwise negatively dependent random variables. J. Inequal. Appl. 2012., 2012: Article ID 50

    Google Scholar 

  32. Wu QY: Probability Limit Theory for Mixed Sequence. China Sci. & Technol. Press, Beijing; 2006.

    Google Scholar 

  33. Sung SH: Moment inequalities and complete moment convergence. J. Inequal. Appl. 2009., 2009: Article ID 271265

    Google Scholar 

  34. Chen PY, Volodin A: Limiting behaviour of moving average processes under negative association assumption. Theory Probab. Math. Stat. 2008, 77: 165–176.

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

The authors are most grateful to the editor Andrei Volodin and anonymous referees for careful reading of the manuscript and valuable suggestions which helped in significantly improving an earlier version of this paper. The work was supported by the National Natural Science Foundation of China (11171001, 11201001), Doctoral Research Start-up Funds Projects of Anhui University and Natural Science Foundation of Anhui Province (1308085QA03, 1208085QA03).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Xiaoqin Li.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

All authors read and approved the final manuscript.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License ( https://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Wang, X., Shen, A. & Li, X. A note on complete convergence of weighted sums for array of rowwise AANA random variables. J Inequal Appl 2013, 359 (2013). https://doi.org/10.1186/1029-242X-2013-359

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/1029-242X-2013-359

Keywords