Abstract
We introduce the fractional mixed fractional Brownian sheet and investigate the small ball behavior of its sup-norm statistic by establishing a general result on the small ball probability of the sum of two not necessarily independent joint Gaussian random vectors. Then, we state general conditions and characterize the sufficiency part of the lower classes of some statistics of the above process by an integral test. Finally, when we consider the sup-norm statistic, the necessity part is given by a second integral test.
1. Introduction
Let be a fractional Brownian motion (FBM) with index , and a fractional Brownian sheet (FBS) with index . We refer to [1] for further information about the FBM and the FBS. Denote by and two real numbers such that .
Define a fractional mixed fractional Gaussian process by a suitable combination of some appropriate fractional Gaussian processes. In the sequel, we consider the following three examples.
Example 1.1. The fractional mixed fractional Brownian motion (FMFBM) is defined by where and are independent FBM with .
Example 1.2. The fractional mixed fractional Brownian motion and fractional Brownian sheet (FMFBMFBS) are defined by where the FBM and the FBS are independent.
Example 1.3. The fractional mixed fractional Brownian sheet (FMFBS) is defined by where and are independent FBS with .
The motivation supporting this paper is threefold.
(i)The first goal of the FMFBS deals with the potential applications. Since the FMFBM, the FMFBMFBS, and the FMFBS can be analyzed based on the large bodies of knowledge on FBM and FBS, it can be used in the same fields, that is, natural time series in economics, fluctuations in solid, hydrology, and, more recently, by new problems in mathematical finance, telecommunication networks, and the environment (see [2–4]).(ii)A second application deals with the small ball probability problem of the sum of two not necessarily joint centered Gaussian random vectors and in a separable Banach space with norm (see [5]). The small ball behavior of the FMFBS under the uniform norm can be investigated as a special case of the small ball probability problem of the sum of two centered Gaussian random vectors, having a log-type small ball factor (see [6]).(iii)Last but not least, this article extends El-Nouty's results [6–9] and consequently answers some new questions. Recall first two definitions of the Lévy classes, stated in [10]. Let be a stochastic process defined on the basic probability space .
Deffinition 1.4. The function belongs to the lower-lower class of the process if, for almost all , there exists such that for every .
Deffinition 1.5. The function , belongs to the lower-upper class of the process , if, for almost all , there exists a sequence with , as , such that .
In the spirit of [6–9, 11], the main aim of this paper is to characterize the lower classes of the uniform norm of the FMFBS for any . More precisely, we want to compare the influence of two FBSs and to measure the weight of a log-type small ball factor versus another one.
2. Main Results
Our first result is given in the following theorem.
Theorem 2.1. Let and be any two joint Gaussian random vectors in a
separable Banach space with norm .
Assume that there exist and such that one has, for any small enough,
with
and .
If or ( and ),
then there exists depending on only such that one has, for any small enough,
Since the study of the lower classes of the FMFBM (resp., FMFBMFBS) under the uniform norm was investigated in [8] (resp., [9]), we focus our attention to the FMFBS. Set Note first that, by the scaling property, we have, for any , where is named the small ball function and the scaling factor.
Recall that the small ball behavior of the FBS under the uniform norm was studied in [12, 13].
Set , which is in . We introduce the number taking its values in . As a direct consequence of Theorem 2.1, we have the following corollary.
Corollary 2.2. There is a constant , depending on and only, such that one has, for any small enough,
Recall that we suppose . In the sequel, there is no loss of generality to assume also that Thus when ( and ), ( and ), or ( and ), we emphasize that , that is, we have a log-type small ball factor.
Note first that the minimum plays a key role. This is not really surprising. Indeed, this phenomenon was already observed in [8, 9].
It appears that the sufficiency part of the lower classes of can be stated in a general framework. Roughly speaking, we follow the same lines as those of [6, 7].
Let be a real-valued statistic of the two independent FBSs, and , such that is a nondecreasing function of .
The following notation is needed. If is a Hausdorff compact space, we denote, by , the space of all continuous functions from to equipped with the classical sup-norm. Let be the product space equipped with the product topology. Denote, by L , the Gaussian measure associated to and and defined on , the Borel -field of .
We assume that satisfies the following three conditions:
(i)The scaling condition. There exists such that (ii)The convexity condition. There exists a convex and -measurable function : such that, for any , and , with probability 1.(iii)The log-type small ball condition. There exist and a constant , depending on and only such that we have, for any small enough,
Note that these conditions generalize those of [6, 7]. The small ball function still plays a key role. The convexity of the function defined by , is ensured by (ii) (see [14, 15]).
Our second result is given in the following theorem.
Theorem 2.3. Let be a positive nondecreasing function of .
Assume that there exists such that .
If
then one has
The sup-norm statistic clearly satisfies the three above conditions with and . Now, we characterize the necessity part of the lower classes of the FMFBMFBS. Our main result is stated in the following theorem.
Theorem 2.4. Let be a positive nondecreasing function of such that is a nonincreasing function of
If
then we have
First, we can notice that Theorem 2.3 involves and . If , Theorem 2.3 looks like [7, Theorem 1] or else like [6, Theorem 1.1]. Theorem 2.4 has the same form as the necessity part of [11, Theorem 1.2] or as the theorems obtained in [6–9]. In his previous works on the study of the lower classes, the author showed that the methodology in [11] led to two integral tests, these tests are actually identical when and . Here, . This is why the integral tests of Theorems 2.3 and 2.4 have different forms. Moreover, since , we must assume, as in [6–9], that is not only bounded, but also a nonincreasing function of . This last assumption will play a key role in some proofs. Finally, although they are two different integral tests, Theorems 2.3 and 2.4 are sharp. Indeed, set, if , or else (i.e., ) If is small enough, then Theorem 2.3 yields , or else if is large enough, then by applying Theorem 2.4.
In Section 3, we prove Theorem 2.1. The proof of Theorem 2.4 is postponed to Sections 4 and 5. In the latter, we establish some key small ball estimates. Note also that these estimates can be of independent interest. The proofs, which are modifications of those of [6, 7], will be consequently omitted, in particular, the proof of Theorem 2.3.
3. Proof of Theorem 2.1
Recall, first, a Gaussian correlation inequality stated in [5].
Theorem A. Let be a centered Gaussian measure on a separable Banach space . Then for any , and any symmetric convex sets and in , In particular,
Rougly speaking, the proof follows the same lines as those in [15] and will be split into two parts: the lower bound and the upper one.
Part I. The Lower Bound
Theorem A implies, for any and ,
Then we get, by using (2.1), Hence, since () or ( and ), there exists depending on only such that we have, for any small enough, and the lower bound follows by taking and .
Part II. The Upper Bound
A new use of Theorem A implies, for any and ,
Then combining (2.1) with the fact that () or ( and ), there exists depending on only such that we have, for any small enough, and the upper bound follows by taking and .
By choosing , we complete the proof of Theorem 2.1.
Remark 3.1. When and are independent, there is a simple proof without using the correlation inequality in the spirit of [5]. A direct proof of Corollary 2.2 can also be done as in [9].
4. Proof of Theorem 2.4, Part I
To simplify the reading of our paper, we introduce the following notation. Set and .
Suppose here that, with probability 1, for all large enough. We want to prove that and .
In the sequel, there is no loss of generality to assume that is a continuous function of .
Lemma 4.1. One has
To prove Theorem 2.4, we will show that when and .
Following the same lines as those in [11], our aim is to construct a suitable subset of such that we have the following property for an appropriate family of sets in a basic probability space: given , there exist a number and an integer such that
Lemma 4.2. When and we can find a sequence with the two following properties:
Remark 4.3. The condition “ is a nonincreasing function of ” is essential to prove Lemma 4.2 (see [7, page 373]).
To continue the construction of the set , we need the following definition and notation.
Definition 4.4. Consider the interval . If then one notes
Next, set which is finite by Lemma 4.1 and where was defined in Corollary 2.2.
Notation
(i)(ii)
( depends on and only);(iii)
where is fixed, and ;(iv)
Now, we can define the set as follows: Since it is assumed that is a nonincreasing function of (being a particular case of the condition “ is bounded”), Moreover, since , we get . Hence is always an empty set (by construction). Thus we obtain .
We have, by Lemma 4.2,
Lemma 4.5. such that card , one has
Proof. Set and where .
We have
Note that, when ,
we have .
Moreover, since by hypothesis, (4.9) implies
when hence are large enough.
Thus, since ,
(4.10) implies (4.8).
The proof of Lemma 4.5 is now complete.
5. Proof of Theorem 2.4, Part II
Consider, now, the events . We have directly , and, by (4.7), . To prove (4.2), we remark that, given can be rewritten as follows: , where , and .
Our first key small ball estimate is given in the following lemma.
Lemma 5.1. Consider , and . Then one has where depends on and only.
Proof. Set and .
We have
Denote, by ,
the integer part of a real .
Let .
We consider the sequence ,
where and .
Consider also the rectangles ,
where .
Their area is .
Let be the event defined by
We have .
Moreover, we
have also
where
Before
rewritting ,
we recall the integral representation of ,
given by
where ,
is a standard Brownian sheet, ,
and is a normalizing constant.
Hence can be rewritten by (5.6) as follows: ,
where
Note also that and are independent.
Since is maximum at and and are independent, we have
The integral
representation of implies that and
Denote, by ,
the distribution function of the absolute value of a standard Gaussian random
variable. Then we obtain
and therefore, .
Choosing ,
we get .
Lemma 5.1 is proved.
Lemma 5.2. and where and are numbers.
Proof. Setting and ,
Lemma 5.1 implies
Consider, first,
the case when .
Lemma 4.2 implies that, for all ,
we have .
Then we can establish
Combining (5.11)
with (5.12), we get
which is the
first part of Lemma 5.2.
Consider, now,
the case .
Combining (5.11)
with the definition of ,
we have
Since ,
we get
and
consequently, by noting that and by assuming ,
we have
Lemma 5.2 is,
therefore, proved.
To deal with the set , we first state a standard large deviation result and a technical lemma (see [6]).
Lemma A. Let be a separable real-valued centered Gaussian process such that with probability 1 and satisfying, for any where Then one has, for where is a positive constant independent of and
Lemma B. One has, for where is small enough, where .
Building on Lemmas A and B, one can establish our last key small ball estimate in the following result.
Lemma 5.3. Let be a real number such that . Set Then one has, for where depend on only, depend on only, is defined as in Lemma B, and are defined as in Lemma A.
Proof. Set .
Set . If then and .
Based on (5.6), and can be split as follows:
where we have,
for (1,2),(3,4),
Note that and are independent as and .
Equation (5.23) implies that the FMFBS can be rewritten as follows: , where
Set
Then, given ,
we have (see [11])
Equation (5.20) implies
and, consequently,
If we choose ,
then we get the first term of the RHS of Lemma 5.3.
Next, we want
to obtain an upper bound of
where we have, for (1,2),(3,4),
We can show, by
standard computations, that
Denote, by ,
the covariance function of a FBM .
Set ,
the covariance function of the process defined by
where is a Wiener process.
Since
we have, for any
Consider II
first. We get, by the inequality of Cauchy-Schwarz,
Consider I now.
A straight computation implies that there exists depending on such that
and, consequently, by the inequality of Cauchy-Schwarz,
So we get
Hence,
combining (5.35) with (5.36) and (5.39), we have
An application of Lemma A with
, and implies that
Similarly, we can establish
Set .
Recall that and .
Combining (5.32) with (5.41) and (5.42), we get
that is the
second term of the RHS of Lemma 5.3.
Finally, we
can establish a similar result for ,
that is,
which achieves the proof of Lemma 5.3.
Finally, we state the last technical lemma.
Lemma 5.4. There exists an integer such that if then for given one has .
Proof. Let and .
We have, by Lemma 4.5,
Let .
Then and .
Thus we have .
Set and .
Note that , and .
By using Lemma 5.3 and letting ,
we end the proof of Lemma 5.4.
Lemmas 5.2 and 5.4 yield that (4.2) holds. Combining Borel-Cantelli's second lemma with (4.2) and (4.7), we show that, given , and, consequently, . The proof of Theorem 2.4 is now complete.
Acknowledgment
The author thanks the referee for the insightful comments.