Hindawi Publishing Corporation

Abstract and Applied Analysis

Volume 2012, Article ID 572493, 13 pages
doi:10.1155/2012/572493

Research Article
Convergence Rates in the Strong Law of Large
Numbers for Martingale Difference Sequences

Xuejun Wang, Shuhe Hu, Wenzhi Yang, and Xinghui Wang

School of Mathematical Science, Anhui University, Hefei 230039, China

Correspondence should be addressed to Xinghui Wang, wangxinghuial@163.com

Received 31 May 2012; Accepted 14 June 2012

Academic Editor: Sung Guen Kim

Copyright q 2012 Xuejun Wang et al. This is an open access article distributed under the Creative
Commons Attribution License, which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.

We study the complete convergence and complete moment convergence for martingale difference
sequence. Especially, we get the Baum-Katz-type Theorem and Hsu-Robbins-type Theorem for
martingale difference sequence. As a result, the Marcinkiewicz-Zygmund strong law of large
numbers for martingale difference sequence is obtained. Our results generalize the corresponding
ones of Stoica 2007, 2011.

1. Introduction
The concept of complete convergence was introduced by Hsu and Robbins 1 as follows.
A sequence of random variables {Un , n ≥ 1} is said to converge completely to a constant C if
∞
n1 P {|Un − C| > ε} < ∞ for all ε > 0. In view of the Borel-Cantelli lemma, this implies
that Un → C almost surely a.s.. The converse is true if the {Un , n ≥ 1} are independent.
Hsu and Robbins 1 proved that the sequence of 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 2 proved the converse. The result of Hsu-
Robbins-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
is Baum and Katz 3 for the strong law of large numbers as follows.

Theorem A see Baum and Katz 3. Let α > 1/2 and let αp > 1. Let {Xn , n ≥ 1} be a sequence
of independent and identically distributed random variables. Assume further that EX1  0 if α ≤ 1.
Then the following statements are equivalent:
i E|X1 |p < ∞,
 
ii ∞ n1 n
αp−2
P max1≤k≤n | ki1 Xi | > εnα  < ∞ for all ε > 0.
2 Abstract and Applied Analysis

Motivated by Baum and Katz 3 for independent and identically distributed
random variables, many authors studied the Baum-Katz-type Theorem for dependent
random variables; see, for example, ϕ-mixing random variables, ρ-mixing random variables,
negatively associated random variables, martingale difference sequence, and so forth.
Our emphasis in the paper is focused on the Baum-Katz-type Theorem for martingale
difference sequence. Recently, Stoica 4, 5 considered the following series that describes the
rate of convergence in the strong law of large numbers:
  

∞  n 
 
nαp−2 P α
 Xi  > εn . 1.1
n1
 i1 

They obtained the follow results.

Theorem B see Stoica 4. Let {Xn , n ≥ 1} be an Lp -bounded martingale difference sequence, and
let 0 < 1/α < 2 < p. Then series 1.1 converges for all ε > 0.

Theorem C see Stoica 5. i Let 1 < p < 2, 1 ≤ 1/α ≤ p and let ε > 0. Then the series 1.1
converges for any martingale difference sequence {Xn , n ≥ 1} bounded in Lp .
ii Let p  α  1 and ε > 0. Then the series 1.1 converges for any martingale difference
sequence {Xn , n ≥ 1} satisfying supn≥1 E|Xn |ln |Xn | < ∞.

The main purpose of the paper is to further study the Baum-Katz-type Theorem for
martingale difference sequence. We have the following generalizations.
i Our results include Baum-Katz-type Theorem and Hsu-Robbins-type Theorem see
Hsu and Robbins 1 as special cases.
ii Our results generalize Theorems B and C for the partial sum to the case of maximal
partial sum.
iii Our results not only generalize Theorem B for 0 < 1/α < 2 < p and Theorem C i
for 1 < p < 2, 1 ≤ 1/α ≤ p to the case of α > 1/2, p > 1 and αp ≥ 1 but also generalize
Theorem C ii for α  1 to the case of α ≥ 1.
Throughout the paper, let {Xn , n ≥ 1} be a sequence of random variables defined on

a fixed probability space Ω, F, P . Denote Sn  ni1 Xi , S0  0, ln x  ln maxx, e, x 
xIx ≥ 0, and F0  {Ω, ∅}. an  bn stands for an  Obn . C, C1 –C4 denote positive constants
which may be different in various places. x denotes the integer part of x. Let IA be the
indicator function of the set A.
Let {Fn , n ≥ 1} be an increasing sequence of σ fields with Fn ⊂ F for each n ≥ 1. If
Xn is Fn measurable for each n ≥ 1, then σ fields {Fn , n ≥ 1} are said to be adapted to the
sequence {Xn , n ≥ 1}, and {Xn , Fn , n ≥ 1} is said to be an adapted stochastic sequence.

Definition 1.1. If {Xn , Fn , n ≥ 1} is an adapted stochastic sequence with

EXn | Fn−1   0 a.s. 1.2

and E|Xn | < ∞ for each n ≥ 1, then the sequence {Xn , Fn , n ≥ 1} is called a martingale
difference sequence.

The following two definitions will be used frequently in the paper.

Abstract and Applied Analysis 3

Definition 1.2. A real-valued function lx, positive and measurable on 0, ∞, is said to be
slowly varying if

lxλ
lim 1 1.3
x → ∞ lx

for each λ > 0.

Definition 1.3. A sequence {Xn , n ≥ 1} of random variables is said to be stochastically

dominated by a random variable X if there exists a positive constant C, such that

P |Xn | > x ≤ CP |X| > x 1.4

for all x ≥ 0 and n ≥ 1.

Our main results are as follows.

Theorem 1.4. Let α > 1/2, p > 1 and let αp ≥ 1. Let {Xn , Fn , n ≥ 1} be a martingale difference
sequence, which is stochastically dominated by a random variable X. Let lx > 0 be a slowly varying
function as x → ∞. Supposing that supi≥1 EXi2 | Fi−1  ≤ C a.s. if p ≥ 2 and
 
E|X|p l |X|1/α < ∞, 1.5

then for any ε > 0,


∞

 
nαp−2 lnP maxSj  ≥ εnα < ∞. 1.6
1≤j≤n
n1

Theorem 1.5. Let α > 1/2, p > 1 and let αp > 1. Let {Xn , Fn , n ≥ 1} be a martingale difference
sequence, which is stochastically dominated by a random variable X. Let lx > 0 be a slowly varying
function as x → ∞. Supposing that supi≥1 E|Xi |2 | Fi−1  ≤ C a.s. if p ≥ 2 and 1.5 holds, then for
any ε > 0,
   
∞  Sj 
nαp−2
lnP sup α  ≥ ε < ∞. 1.7
n1 j≥n j

For p  1 and lx  1, we have the following theorem.

Theorem 1.6. Let α ≥ 1, and let {Xn , Fn , n ≥ 1} be a martingale difference sequence, which is
stochastically dominated by a random variable X. Supposing that

E|X|ln |X| < ∞, 1.8

then for any ε > 0,


∞

 
nα−2 P maxSj  ≥ εnα < ∞. 1.9
1≤j≤n
n1
4 Abstract and Applied Analysis

The following theorem presents the complete moment convergence for martingale
difference sequence.

Theorem 1.7. Letting the conditions of Theorem 1.4 hold, then for any ε > 0,


∞

 
αp−2−α
n lnE maxSj  − εnα < ∞. 1.10
1≤j≤n
n1

Remark 1.8. If we take lx ≡ 1 in Theorem 1.4, then we can not only get the Baum-Katz-
type Theorem for martingale difference sequence but also consider the case of pα  1.
Furthermore, if we take lx ≡ 1, α  1, and p  2 in Theorem 1.4, then we can get the
Hsu-Robbins-type Theorem see Hsu and Robbins 1 for martingale difference sequence.

Remark 1.9. As stated above, our Theorems 1.4 and 1.5 not only generalize the corresponding
results of Theorems B and C for the partial sum to the maximal partial sum but also expand
the scope of α and p.

Remark 1.10. If we take lx ≡ 1 in Theorem 1.4, then we can get the Marcinkiewicz-Zygmund
strong law of large numbers for martingale difference sequence as follows:

1 n
Xi −→ 0, a.s. 1.11
nα i1

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

Lemma 2.1 see 6, Theorem 2.11. If {Xi , Fi , 1 ≤ i ≤ n} is a martingale difference and q > 0,
then there exists a constant C depending only on p such that

  q  ⎧   ⎫
 k  ⎨ n   q/2
⎬
 
E max  Xi  ≤C E E Xi2 | Fi−1  E max|Xi |q . 2.1
1≤k≤n  ⎩ 1≤i≤n ⎭
i1 i1

Lemma 2.2. Let {Xn , n ≥ 1} be a sequence of random variables, which is stochastically dominated by
a random variable X. Then for any a > 0 and b > 0, the following two statements hold:
 
E |Xn |a I|Xn | ≤ b ≤ C1 EX a I|X| ≤ b  ba P |X| > b,
    2.2
E |Xn |a I|Xn | > b ≤ C2 E |X|a I|X| > b ,

where C1 and C2 are positive constants.

Lemma 2.3 cf. 7. If lx > 0 is a slowly varying function as x → ∞, then
i limx → ∞ ltx/lx  1 for each t > 0; limx → ∞ lx  u/lx  1 for each u ≥ 0,
ii limk → ∞ sup2k ≤x<2k1 lx/l2k   1,
iii limx → ∞ xδ lx  ∞, limx → ∞ x−δ lx  0 for each δ > 0,
Abstract and Applied Analysis 5

iv C1 2kr lε2k  ≤ kj1 2jr lε2j  ≤ C2 2kr lε2k  for every r > 0, ε > 0, positive integer k and
some C1 > 0, C2 > 0,

v C3 2kr lε2k  ≤ ∞jk 2 lε2  ≤ C4 2 lε2  for every r < 0, ε > 0, positive integer k and
jr j kr k

some C3 > 0, C4 > 0.

3. Proofs of the Main Results

Proof of Theorem 1.4. For fixed n ≥ 1, denote

Yni  Xi I|Xi | ≤ nα  − EXi I|Xi | ≤ nα  | Fi−1 , i  1, 2, . . . . 3.1

Since Xi  Xi I|Xi | > nα   Yni  EXi I|Xi | ≤ nα  | Fi−1 , we can see that


∞

 
nαp−2 lnP maxSj  ≥ εnα
1≤j≤n
n1
  j  

∞   εnα
 α 
≤ nαp−2 lnP max Xi I|Xi | > n  ≥
1≤j≤n  3
n1 i1
  j  

∞   εnα 3.2
 
 nαp−2 lnP max EXi I|Xi | ≤ n  | Fi−1  ≥
α
1≤j≤n  3
n1 i1
  j  

∞   εnα
 
 nαp−2 lnP max Yni  ≥
1≤j≤n  3
n1 i1

: H  I  J.

For H, we have by Markov’s inequality, Lemma 2.2, and 1.5 that

  j 
∞  
 α 
H nαp−2−α
lnE max Xi I|Xi | > n 
1≤j≤n 
n1 i1


∞ 
n
≤ nαp−2−α ln E|Xi |I|Xi | > nα 
n1 i1


∞
 nαp−1−α lnE|X|I|X| > nα 
n1


∞ 
∞   
 nαp−1−α ln E |X|I m < |X|1/α ≤ m  1
n1 mn

∞ 
  
m
 E |X|I m < |X|1/α ≤ m  1 nαp−1−α ln
m1 n1

∞ 
  log
2 m 1 
2i
1/α
≤ E |X|I m < |X| ≤m1 nαp−1−α ln
m1 i1 n2i−1
6 Abstract and Applied Analysis

∞ 
  log
2 m 1  
 E |X|I m < |X|1/α ≤ m  1 2iαp−1 l 2i
m1 i1


∞     
 E |X|I m < |X|1/α ≤ m  1 2log2 m 1αp−1 l 2log2 m 1
m1

∞ 
  
 E |X|I m < |X|1/α ≤ m  1 mαp−1 lm
m1
 
 E|X|p l |X|1/α < ∞.
3.3

For I, we have by Markov’s inequality and 3.3 that

 j  

∞  
 
I nαp−2−α lnE max EXi I|Xi | ≤ nα  | Fi−1 
1≤j≤n 
n1 i1
  j 
∞  
 
 nαp−2−α
lnE max EXi I|Xi | > n  | Fi−1 
α
1≤j≤n 
n1 i1 3.4

∞ 
n
≤ nαp−2−α ln E|Xi |I|Xi | > nα 
n1 i1
 
 E|X|p l |X|1/α < ∞.

To prove 1.6, it suffices to show that

  j  

∞   εnα
  3.5
J : nαp−2
lnP max Yni  ≥ < ∞.
1≤j≤n  3
n1 i1

For fixed n ≥ 1, it is easily seen that {Yni , Fi , i ≥ 1} is still a martingale difference. By Markov’s
inequality and Lemma 2.1, we have that for any q ≥ 2,

 j 

∞   q
 
J nαp−2−αq
lnE max Yni 
1≤j≤n 
n1 i1
 

∞ 
n 
∞ n   q/2 3.6
q
 nαp−2−αq
ln E|Yni |  nαp−2−αq
lnE E Yni | Fi−1
2

n1 i1 n1 i1

: J1  J2 .

We consider the following three cases.

Case 1 αp > 1 and p ≥ 2. Take q large enough such that q > maxp, αp − 1/α − 1/2,
which implies that αp − 2 − αq  q/2 < −1.
Abstract and Applied Analysis 7

For J1 , we have by Cr ’s inequality, Lemma 2.2, 3.3, Lemma 2.3, and 1.5 that

∞ 
n
 
J1  nαp−2−αq ln E |Xi |q I|Xi | ≤ nα 
n1 i1


∞ 
n
  
∞ 
n
 nαp−2−αq ln E |X|q I|X| ≤ nα   nαp−2−αq ln nαq P |X| > nα 
n1 i1 n1 i1


∞
  
∞
 nαp−1−αq lnE |X|q I|X| ≤ nα   nαp−1 lnP |X| > nα 
n1 n1


∞
  
∞
≤ nαp−1−αq lnE |X|q I|X| ≤ nα   nαp−1−α lnE|X|I|X| > nα 
n1 n1


∞
 
 nαp−1−αq lnE |X|q I|X| ≤ nα 
n1


∞ 
n  
≤ nαp−q−1 ln j αq P j − 1 < |X|1/α ≤ j
n1 j1

  3.7

∞
1/α
∞
 j P j − 1 < |X|
αq
≤j nαp−q−1 ln
j1 nj


∞   
∞ 
2i1
1/α
≤ j P j − 1 < |X|
αq
≤j nαp−q−1 ln
j1 ilog2 j n2i


∞    ∞  
 j αq P j − 1 < |X|1/α ≤ j 2iαp−q l 2i
j1 ilog2 j


∞    
 j αq P j − 1 < |X|1/α ≤ j j αp−q l j
j1


∞
   
 j αp l j P j − 1 < |X|1/α ≤ j
j1
 
 E|X|p l |X|1/α < ∞.

Note that supi≥1 EXi2 | Fi−1  ≤ C, a.s. if p ≥ 2. We have by Lemma 2.3 that
 

∞ 
n   q/2
J2 ≤ nαp−2−αq
lnE E Xi I|Xi | ≤ n  | Fi−1
2 α

n1 i1
 q/2

∞ 
n  
≤ nαp−2−αq lnE supE Xi2 | Fi−1 3.8
n1 i1 i≥1


∞
 nαp−2−αqq/2 ln < ∞.
n1
8 Abstract and Applied Analysis

Case 2 αp > 1 and p < 2. Take q  2. Similar to the proof of 3.6 and 3.7, we can get that


∞ 
n  
J nαp−2−2α ln E Xi2 I|Xi | ≤ nα  < ∞. 3.9
n1 i1

Case 3 αp  1. Note that p  1/α < 2. Take q  2, and similar to the proof of 3.9, we still
have J < ∞.
From the statements mentioned previously, we have proved 3.5. This completes the
proof of the theorem.

Proof of Theorem 1.5. We have by Lemma 2.3 that

       

∞  Sj  ∞ 2 m
−1  Sj 
αp−2
n lnP  
sup α  > ε  nαp−2  
lnP sup α  > ε
n1 j≥n j m1 n2m−1 j≥n j
    m

∞  Sj  2−1
 P sup  α  > ε 2mαp−2 l2m 
m1 j≥2m−1 j n2m−1
   

∞  Sj 
 2 mαp−1
l2 P
m  
sup  α  > ε
m1 j≥2m−1 j
   

∞  Sj 
 2 l2 P sup max  α  > ε
mαp−1 m
k≥m 2 ≤j<2 j
k−1 k
m1
 
∞ ∞  
≤ 2mαp−1 
l2  P max Sj > ε2
m  αk−1 3.10
m1 km 1≤j≤2k

 

∞   
k
 P max Sj  > ε2αk−1 2mαp−1 l2m 
k1 1≤j≤2k m1
 

∞    
 2 kαp−1
l 2k P max Sj  > ε2αk−1
k1 1≤j≤2k

 −1
∞ 2k1   ε α
 nαp−2  
lnP max Sj > α n
k1 n2k
1≤j≤n 4


∞   ε α
 n αp−2  
lnP max Sj > α n .
n1
1≤j≤n 4

The desired result 1.7 follows from the inequality above and 1.6 immediately.

Proof of Theorem 1.6. We use the same notation as that in Theorem 1.4. According to the proof
of Theorem 1.4, we can see that J < ∞ for p  1 and lx  1 under the conditions of
Theorem 1.6. So it suffices to show that H < ∞ and I < ∞ for p  1 and lx  1.
Abstract and Applied Analysis 9

Similar to the proof of 3.3, we have


∞
H n−1 E|X|I|X| > nα 
n1


∞ 
∞   
 n−1 E |X|I m < |X|1/α ≤ m  1
n1 mn

∞ 
  
m
3.11
 E |X|I m < |X|1/α ≤ m  1 n−1
m1 n1


∞   
 E |X|I m < |X|1/α ≤ m  1 ln m
m1

 E|X|ln |X| < ∞.

Similar to the proof of 3.4 and 3.11, we can get that


∞ 
n
I n−2 E|Xi |I|Xi | > nα 
n1 i1
3.12

∞
−1
 n E|X|I|X| > n  < ∞. α

n1

This completes the proof of the theorem.

Proof of Theorem 1.7. For any ε > 0, we have by Theorem 1.4 that


∞

 
nαp−2−α lnE maxSj  − εnα
1≤j≤n
n1


∞ ∞

 
 nαp−2−α
ln P maxSj  − εnα > t dt
0 1≤j≤n
n1


∞  nα

 
 nαp−2−α ln P maxSj  − εnα > t dt
0 1≤j≤n
n1
∞
3.13

∞  
 nαp−2−α ln P maxSj  − εnα > t dt
nα 1≤j≤n
n1


∞

∞ ∞

   
≤ nαp−2 lnP maxSj  > εnα  nαp−2−α ln P maxSj  > t dt
1≤j≤n nα 1≤j≤n
n1 n1


∞ ∞

 
 nαp−2−α ln P maxSj  > t dt.
nα 1≤j≤n
n1
10 Abstract and Applied Analysis

Hence, it suffices to show that


∞ ∞

 
Q : nαp−2−α
ln P maxSj  > t dt < ∞. 3.14
nα 1≤j≤n
n1

For t > 0, denote

Zti  Xi I|Xi | ≤ t − EXi I|Xi | ≤ t | Fi−1 , i  1, 2, . . . . 3.15

Since Xi  Xi I|Xi | > t  Zti  EXi I|Xi | ≤ t | Fi−1 , it follows that

∞   j  

∞   t
 
Q≤ αp−2−α
n ln P max Xi I|Xi | > t > dt
nα 1≤j≤n  3
n1 i1

∞   j  

∞   t
 
 n αp−2−α
ln P max EXi I|Xi | ≤ t | Fi−1  > dt
nα 1≤j≤n  3 3.16
n1 i1

∞   j  

∞   t
 
 nαp−2−α ln P max Zti  > dt
nα 1≤j≤n  3
n1 i1

: Q1  Q2  Q3 .

Similar to the proof of 3.3, we have by Markov’s inequality and Lemma 2.2 that

∞   j 
∞  
 
Q1  nαp−2−α ln t−1 E max Xi I|Xi | > t dt
n1 nα 1≤j≤n i1



∞ ∞
 nαp−1−α
ln t−1 E|X|I|X| > tdt
n1 nα


∞ ∞  m1

α

 nαp−1−α
ln t−1 E|X|I|X| > tdt
n1 mn mα


∞ 
∞
 nαp−1−α ln m−1 E|X|I|X| > mα  3.17
n1 mn


∞ 
m
 m−1 E|X|I|X| > mα  nαp−1−α ln
m1 n1


∞
 m−1 E|X|I|X| > mα mαp−α lm
m1


∞
 nαp−1−α lnE|X|I|X| > nα  < ∞.
n1
Abstract and Applied Analysis 11

According to the proof of 3.17, we have by Markov’s inequality and Lemma 2.2 that

∞  j 

∞  
−1 
Q2  nαp−2−α
ln t E max EXi I|Xi | ≤ t | Fi−1  dt
nα 1≤j≤n 
n1 i1

∞   j 
∞  
−1   3.18
 nαp−2−α
ln t E max EXi I|Xi | > t | Fi−1  dt
n1 nα 1≤j≤n i1



∞ ∞
 nαp−1−α
ln t−1 E|X|I|X| > tdt < ∞.
n1 nα

For any t > 0, it is easily seen that {Zti , Fi , i ≥ 1} is still a martingale difference. By Markov’s
inequality and Lemma 2.1, we have that for any q ≥ 2,

 j  ⎞ ⎛

∞ ∞ q
 
Q3  nαp−2−α ln t−q E⎝max Zti  ⎠dt
nα 1≤j≤n 
n1 i1


∞ ∞ 
n
 n αp−2−α
ln t−q E|Zti |q dt
n1 nα i1 3.19
∞  

∞ 
n   q/2
−q
 nαp−2−α
ln t E E Zti | Fi−1
2
dt
n1 nα i1

: Q31  Q32 .

We still consider the following three cases.

Case 1 αp > 1 and p ≥ 2. Take q large enough such that q > maxp, αp − 1/α − 1/2,
which implies that αp − 2 − αq  q/2 < −1. We have by Lemma 2.2 and 3.17 that


∞ ∞ 
n
 
Q31  nαp−2−α ln t−q E |Xi |q I|Xi | ≤ t dt
n1 nα i1


∞ ∞
 
 nαp−1−α ln t−q E |X|q I|X| ≤ t dt
n1 nα


∞ ∞
 nαp−1−α ln P |X| > tdt
n1 nα
∞ 3.20

∞
 
−q
≤ nαp−1−α
ln t E |X|q I|X| ≤ t dt
n1 nα


∞ ∞
 nαp−1−α ln t−1 E|X|I|X| > tdt
n1 nα


∞ ∞
 
 nαp−1−α ln t−q E |X|q I|X| ≤ t dt.
n1 nα
12 Abstract and Applied Analysis

Hence, similar to the proof of 3.7, we can see that


∞ ∞  m1

α
 
Q31  nαp−1−α
ln t−q E |X|q I|X| ≤ t dt
n1 mn mα


∞ 
∞
  
≤ nαp−1−α ln mα−1−αq E |X|q I |X| ≤ m  1α
n1 mn


∞
  
m
 mα−1−αq E |X|q I |X| ≤ m  1α nαp−1−α ln
m1 n1


∞
  
 mα−1−αq E |X|q I |X| ≤ m  1α mαp−α lm
m1


∞
   3.21
 nαp−1−αq lnE |X|q I |X| ≤ n  1α
n1


∞
  
 nαp−1−αq lnE |X|q I nα < |X| ≤ n  1α
n1


∞
 
 nαp−1−αq lnE |X|q I|X| ≤ nα 
n1


∞       
 n−1 E |X|p l |X|1/α I nα < |X| ≤ n  1α  E|X|p l |X|1/α
n1
 
 E|X|p l |X|1/α < ∞.

Note that supi≥1 EXi2 | Fi−1  ≤ C, a.s. if p ≥ 2. We have by Lemma 2.3 that

∞  

∞ n   q/2
−q
Q32 ≤ nαp−2−α
ln t E E Xi I|Xi | ≤ n  | Fi−1
2 α
dt
n1 nα i1

∞  q/2

∞ 
n  
−q
 nαp−2−α ln t E supE Xi2 | Fi−1 dt
nα i1 i≥1
n1
3.22

∞ ∞
 nαp−2−αq/2
ln t−q dt
n1 nα


∞
 nαp−2−αqq/2 ln < ∞.
n1

Case 2 αp > 1 and p < 2. Take q  2. Similar to the proof of 3.19 and 3.21, we can get that


∞ ∞ 
n
Q3  nαp−2−α ln t−2 E|Zti |2 dt < ∞. 3.23
n1 nα i1
Abstract and Applied Analysis 13

Case 3 αp  1. Note that p  1/α < 2. Take q  2, and similar to the proof of 3.23, we still
have Q3 < ∞.
From the statements mentioned previously, we have proved 3.14. This completes the
proof of the theorem.

Acknowledgments
The authors are most grateful to the Editor Sung Guen Kim and anonymous referees for care-
ful reading of the paper and valuable suggestions which helped in improving an earlier ver-
sion of this paper. This work was supported by the National Natural Science Foundation of
China 11171001, 11126176, Natural Science Foundation of Anhui Province 1208085QA03,
Provincial Natural Science Research Project of Anhui Colleges KJ2010A005, Doctoral
Research Start-up Funds Projects of Anhui University, the Academic Innovation Team of
Anhui University KJTD001B, and The Talents Youth Fund of Anhui Province Universities
2010SQRL016ZD, 2011SQRL012ZD.

References
1 P. L. Hsu and H. Robbins, “Complete convergence and the law of large numbers,” Proceedings of the
National Academy of Sciences of the United States of America, vol. 33, no. 2, pp. 25–31, 1947.
2 P. Erdös, “On a theorem of Hsu and Robbins,” Annals of Mathematical Statistics, vol. 20, no. 2, pp. 286–
291, 1949.
3 L. E. Baum and M. Katz, “Convergence rates in the law of large numbers,” Transactions of the American
Mathematical Society, vol. 120, no. 1, pp. 108–123, 1965.
4 G. Stoica, “Baum-Katz-Nagaev type results for martingales,” Journal of Mathematical Analysis and Appli-
cations, vol. 336, no. 2, pp. 1489–1492, 2007.
5 G. Stoica, “A note on the rate of convergence in the strong law of large numbers for martingales,”
Journal of Mathematical Analysis and Applications, vol. 381, no. 2, pp. 910–913, 2011.
6 P. Hall, Martingale Limit Theory and Its Application, Academic Presss, New York, NY, USA, 1980.
7 Z. D. Bai and C. Su, “The complete convergence for partial sums of i.i.d. random variables,” Science in
China Series A, vol. 28, no. 12, pp. 1261–1277, 1985.
Advances in Advances in Journal of Journal of
Operations Research
Hindawi Publishing Corporation
Decision Sciences
Hindawi Publishing Corporation
Applied Mathematics
Hindawi Publishing Corporation
Algebra
Hindawi Publishing Corporation
Probability and Statistics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

The Scientific International Journal of

World Journal
Hindawi Publishing Corporation
Differential Equations
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

International Journal of Advances in

Combinatorics
Hindawi Publishing Corporation
Mathematical Physics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Mathematical Problems Abstract and Discrete Dynamics in

Complex Analysis
Hindawi Publishing Corporation
Mathematics
Hindawi Publishing Corporation
in Engineering
Hindawi Publishing Corporation
Applied Analysis
Hindawi Publishing Corporation
Nature and Society
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

International
Journal of Journal of
Mathematics and
Mathematical
Discrete Mathematics
Sciences

Journal of International Journal of Journal of

Hindawi Publishing Corporation Hindawi Publishing Corporation Volume 2014

http://www.hindawi.com
Function Spaces
Hindawi Publishing Corporation
Volume 2014 http://www.hindawi.com http://www.hindawi.com
Stochastic Analysis
Hindawi Publishing Corporation
Optimization
Hindawi Publishing Corporation
Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014