1. Introduction, background, and main results
The study of univariate records is very well developed ( being a classical reference), but that of multivariate records less well so, in part because there are many ways one can formulate the latter concept. See , and the references therein, and [1, Chap. 8] for background.
This paper is mainly about the stochastic process , where is the boundary, or “frontier”, for Pareto records (otherwise known as nondominated records or weak records; consult Definitions 1.1–1.2) in general dimension when the observed sequence of points are assumed (as they are throughout the paper) to be i.i.d. (independent and identically distributed) copies of a
-dimensional random vectorwith independent Exponential coordinates .
Theoretical investigation leading to the results in this paper were spurred by empirical observations whose generation is discussed briefly in Section 5 (see especially Figure 3) and in detail in  and began with the simple result of Theorem 1.4.
Notation: Throughout this paper we abbreviate the th iterate of natural logarithm by and by , and we write for the sum of coordinates of the -dimensional vector .
Unless otherwise specifically noted, all the results of this paper hold for any dimension .
1.1. Pareto records and the record-setting region
We begin with some definitions. Write (respectively, ) to mean that (resp., ) for . (We caution that, with this convention, is weaker than , the latter meaning “ or ”; indeed, but we have neither nor . This distinction will matter little in this paper, since the probability that any coordinate of an observation is repeated or vanishes is , but the distinction is important in .) The notation means , and means .
(a) We say that is a (Pareto) record (or that it sets a record at time ) if for all .
(b) If , we say that is a current record (or remaining record, or maximum) at time if for all .
(c) If , we say that is a broken record at time if it is a record but not a current record, that is, if for all but for some ; in that case, the observation corresponding to the smallest such is said to break or kill the record .
For (or , with the obvious conventions) let denote the number of records with , let denote the number of remaining records at time , and let denote the number of broken records. Note that and are nondecreasing in , but the same is not true for . For dimension , by standard consideration of concomitants [that is, by considering the -dimensional sequence sorted from largest to smallest value of (say) last coordinate] we see that (that is, for dimension , with similar notation used here for ) has, for each , the same (univariate) distribution as ; note, however, the same equality in distribution does not hold for the stochastic processes and .
(a) The record-setting region at time is the (random) closed set of points
(b) We call the (topological) boundary of (relative to the closed positive orthant determined by the origin) its frontier and denote it by .
The terminology in Definition 1.2(a) is natural since the next observation sets a record if and only if it falls in the record-setting region. Note that
and that the current records at time all belong to but lie on its frontier. Observe also that is a closed subset of . Because this paper makes heavy use of the classical probabilistic notion of boundary-crossing probabilities, to avoid confusion we have chosen to use the term “frontier” for , rather than “boundary”, in Definition 1.2(b).
1.2. The record-setting frontier
Our first result shows that deviations of the sum of coordinates for a generic current record at time from are typically of constant order. Observe that the conditional distribution of given that is a current record at time doesn’t depend on ; in particular, it’s the conditional distribution of given that sets a record. Let
be a random variable with that distribution. Letdenote a random variable with the standard Gumbel distribution (i.e., distribution function , ), and write for convergence in law (i.e., in distribution)
This is quite elementary. Let denote the probability that sets a record. Fix
for the moment. Forwe have
and so the conditional density depends on only through . It follows that the density of satisfies
Using the well-known asymptotic equivalence as [see (4.5) below], it is easy to check that, for each fixed , the density of at converges to the standard Gumbel density as . The claimed result thus follows from Scheffé’s theorem (e.g., [4, Thm. 16.12]), which shows that there is in fact convergence in total variation. ∎
This paper primarily concerns the stochastic process , and specifically its “width” as defined next (see Figure 1).
Recall that denotes the frontier of , and let
We define the width of as
Very roughly put, what we will see in this paper is that, unlike of Theorem 1.4, deviations of from are exactly of order ; on the other hand, we will see that deviations of from are of smaller order than . It will follow that the width of the frontier is exactly of order .
We next make some simple observations about the quantities appearing in Definition 1.5 that will prove fundamentally useful to our development.
Lemma 1.6 (characterization of ).
which is nondecreasing in .
The current records at time all belong to , and broken records and non-records all have coordinate-sums (strictly) smaller than some current record. Thus . Conversely, if , then for some ; it follows that . ∎
Lemma 1.7 (two upper bounds on ).
(b) Let . Define
Then, over the event that there are at least remaining records at time , we have
(c) The processes , , and (for any ) all have nondecreasing sample paths.
(a) For , let denote the almost surely unique index such that
Let denote the th coordinate vector. We claim that the points with all belong to (in fact, to ), and then the inequality is immediate. To prove the claim, note that all of the points belong to [because and hence ] but also to [because ].
(b) Over the event , is certainly at most the th-largest sum of coordinates of remaining records, which is in turn at most .
(c) The asserted monotonicity is clear for the bounding processes. The asserted monotonicity of follows easily from the observation that . ∎
It seems difficult to study the processes and bivariately, so we draw all our conclusions about the width process by studying and univariately (that is, separately) and using . The behavior of is well known from classical extreme value theory and is reviewed in Section 2. Conclusions about will be drawn from (i) the upper-bounding processes in Lemma 1.7(a)–(b) together with classical extreme value theory for those bounding processes and (ii) a rather nontrivial lower bound developed in Section 3.
1.3. Main results
We next present the main results of our paper. What the results show, in various precise senses, is that and both concentrate near , with deviations that are , from which it follows of course that . But for we show more, namely, that is the exact scale for , that is, that . We can even narrow things down further: in probability for each , with an almost sure equal to and an almost sure equal to .
Here are our main results for arbitrary but fixed dimension . We consider both convergence in probability (typical behavior) and almost sure largest and smallest deviations from (top and bottom boundary-behavior, respectively) for large .
Theorem 1.8 (Kiefer ).
(a) Typical behavior of :
(b) Top boundaries for :
(c) Bottom boundaries for :
Theorem 1.8 gives rise immediately to the following succinct corollary.
Corollary 1.9 (Kiefer ).
(a) Typical behavior of :
(b) Almost sure behavior for :
In fact, one can show rather simply from Corollary 1.9(b) and the fact that has nondecreasing sample paths that the set (call it ) of limit points of the sequence is almost surely the closed interval . Here is a sketch of the proof. The set is closed, so we need only show that is dense in , which clearly follows if we can show that
the roughly stated idea being that then (a.s.) the sequence “can’t leap downward over any interval i.o.” in its infinitely many downward moves from its to its . To prove (1.3), we first bound from below by , then express the resulting difference with a common denominator, and finally use the consequence of Corollary 1.9(b) to find
who, for the limited purpose of proving a central limit theorem reviewed in Theorem4.1(a) below, “observe that nearly all maxima occur in a thin strip sandwiched between [the] two parallel hyper-planes”
Our results for show that the deviations of from are almost surely negligible on a scale of .
(a) Typical behavior of :
(b) Top outer boundaries for : If , then
(c1) A bottom outer boundary for on the scale of :
(c2) A bottom inner boundary for on the scale of :
Theorem 1.12 gives rise immediately to the following succinct corollary.
(a) Typical behavior of :
(b) Almost sure behavior for : If , then
Consider the process defined at (1.2).
(a) Typical behavior of :
(b) Almost sure behavior for : If , then
and, in particular,
(a) When , at each time there is exactly one current record, is the value of that record, is the closed interval , and .
(c) Theorem 1.14(b) has the following immediate corollary. If, for some positive integer , processes corresponding to dimension , , are defined on a common probability space (regardless of any dependence among the processes), then
That is, roughly speaking, for time large relative to large dimension , the width almost surely concentrates near .
(d) We could have used in the denominators of (1.4), but we chose because of Theorem 1.14(a). A remark of a somewhat similar flavor as (b) for convergence in probability is the following. If, for some integer , processes corresponding to dimension , , are defined on a common probability space (regardless of any dependence among the processes), then
We have not investigated whether this result might extend to dimension growing with .
1.4. Outline of paper
The stochastic process is studied in Section 2, where we prove Theorem 1.8. We treat the process in Section 3, where we prove Theorem 1.12. In Section 4 we assess asymptotic behavior of the record counts , , and introduced following Definition 1.1 as preparation for Section 5, where we produce versions of our main results concerning the record-setting frontier process when time is measured in the number of records (rather than observations ) generated.
2. The process
This section is devoted to the proof of Theorem 1.8 concerning the process defined at (1.1). In light of the characterization provided by Lemma 1.6, Theorem 1.8 follows from results of . Kiefer is concerned with behavior of the law of the iterated logarithm type for the empirical distribution function and sample
-quantiles for a sequence of independent uniformrandom variables, with and , but notes that his results “may easily be translated into results for general laws.” Since we are concerned here with a sequence from the Gamma distribution and with (only) the upper quantile, for completeness and the reader’s convenience we distill Kiefer’s proof(s) for our special case.
Proof of Theorem 1.8.
(a) This is elementary. We have
(b) Kiefer describes two proofs. The first proof observes, for any sequence which is ultimately monotone nondecreasing, that
and applies the Borel–Cantelli lemmas to the sequence of independent events with . The second proof exploits the nondecreasingness of the sample paths of the process noted in Lemma 1.7 and proceeds as follows. If is ultimately monotone nondecreasing and is any strictly increasing sequence of positive integers, then
where we note that the random variables
are independent. Now choose and and apply the Borel–Cantelli lemmas.
(c) For the case of outer-class bottom boundaries, we start with the observation that if is ultimately monotone nondecreasing and is any strictly increasing sequence of positive integers, then
We then choose with and and apply the first Borel–Cantelli lemma.
For the case of inner-class bottom boundaries, we start with the observation that if is ultimately monotone nondecreasing and is any strictly increasing sequence of positive integers, then, recalling the definition (2.1),
We then choose with and with and apply the first Borel–Cantelli lemma to the events and the second Borel–Cantelli lemma to the independent events . ∎
3. The process
3.1. Towards a stochastic lower bound on
To prove Theorem 1.12 we need a stochastic lower bound on to complement the upper bound of Lemma 1.7. For this we use the definitions of the frontier and the closed record-setting region to argue as follows. For , let
denote the open positive orthant determined by . For any set , let denote the number of observations with that fall in . Then
The difficulty with upper-bounding the probability of this event is of course that the last union is uncountable. In the next subsection we produce a geometric lemma whose application effectively bounds the uncountable union by a finite union.
3.2. A geometric lemma
Consider the (uncountable) union of positive orthants whose vertices lie on the hyperplane in where is an integer. We can also form a finite union of positive orthants whose vertices lie on the hyperplane situated a bit further from the origin. Our key geometric lemma guarantees that the uncountable union contains the finite union (see Figure 2).
Given a positive integer , and with
there exists with
and strictly less than the integer , i.e., is at most . Thus we need only (arbitrarily) “sweeten” (i.e., add to) precisely of the entries to obtain with the desired properties. ∎
3.3. A stochastic lower bound on
together with homogeneity [ for and ], that
and so by finite subadditivity
Since the cardinality of equals
we conclude that
where the last inequality holds assuming that as .
We summarize and simplify the bound we have derived in the next proposition, where we assume further that . The bound is the key to the proof of the first assertion in Theorem 1.12(a) and of Theorem 1.12(c1).
Proposition 3.2 (Stochastic lower bound on ).
Let with and . Then
3.4. Proof of Theorem 1.12
In this subsection we prove Theorem 1.12, part by part in the order (a), (c1), (c2), (b).
Proof of Theorem 1.12(a).
Proof of Theorem 1.12(c1).
As noted in Lemma 1.7, the process has nondecreasing sample paths. From this it follows that if is (ultimately) monotone nondecreasing and is any strictly increasing sequence of positive integers, then
To complete the proof, we choose and , bound using Proposition 3.2, and apply the first Borel–Cantelli lemma.
Here are the details. Since and
the hypotheses of Proposition 3.2 are met and
which is summable. ∎
We chose the constant as the coefficient of in parts (a) and (c1) of Theorem 1.12 for convenience. As the proof shows, we could have used any constant larger than .