Partial orders are a special class of relations that play an important role in probability theory.
Basic Theory
Definitions
A partial order on a set is a relation on that is reflexive, anti-symmetric, and transitive. The pair is called a partially ordered set. So for all :
, the reflexive property
If and then , the antisymmetric property
If and then , the transitive property
As the name and notation suggest, a partial order is a type of ordering of the elements of . Partial orders occur naturally in many areas of mathematics, including probability. A partial order on a set naturally gives rise to several other relations on the set.
Suppose that is a partial order on a set . The relations , , , , and are defined as follows:
if and only if .
if and only if and .
if and only if .
if and only if or .
if and only if neither nor .
Note that is the inverse of , and is the inverse of . Note also that if and only if either or , so the relation completely determines the relation . The relation is sometimes called a strict or strong partial order to distingush it from the ordinary (weak) partial order . Finally, note that means that and are related in the partial order, while means that and are unrelated in the partial order. Thus, the relations and are complements of each other, as sets of ordered pairs. A total or linear order is a partial order in which there are no unrelated elements.
A partial order on is a total order or linear order if for every , either or .
Suppose that and are partial orders on a set . Then is an sub-order of , or equivalently is an extension of if and only if implies for .
Thus if is a suborder of , then as sets of ordered pairs, is a subset of . We need one more relation that arises naturally from a partial order.
Suppose that is a partial order on a set . For , is said to cover if but no element satisfies .
If is finite, the covering relation completely determines the partial order, by virtue of the transitive property.
Suppose that is a partial order on a finite set . The covering graph or Hasse graph of is the directed graph with vertex set and directed edge set , where if and only if covers .
Thus, if and only if there is a directed path in the graph from to . Hasse graphs are named for the German mathematician Helmut Hasse. The graphs are often drawn with the edges directed upward. In this way, the directions can be inferred without having to actually draw arrows.
Basic Examples
Of course, the ordinary order is a total order on the set of real numbers . The subset partial order is one of the most important in probability theory:
Suppose that is a set. The subset relation is a partial order on , the power set of .
Details:
We proved this result in the section on sets. To review, recall that for , means that implies . Also means that if and only if . Thus
and if and only if
and imply
Here is a partial order that arises naturally from arithmetic.
Let denote the division relation on the set of positive integers . That is, if and only if there exists such that . Then
is a partial order on .
is a sub-order of the ordinary order .
Details:
Clearly for , since , so is reflexive. Suppose and , where . Then there exist such that and . Substituting gives , and hence . Thus so is antisymmetric. Finally, suppose and , where . Then there exists such that and . Substituting gives , so . Thus is transitive.
If and , then there exists such that . Since , .
The set of functions from a set into a partial ordered set can itself be partially ordered in a natural way.
Suppose that is a set and that is a partially ordered set, and let denote the set of functions . The relation on defined by if and only for all is a partial order on .
Details:
Suppose that .
for all , so .
If and then and for all . Hence for all so .
If and then and for all . Hence for all so .
Note that we don't need a partial order on the domain .
Basic Properties
The proofs of the following basic properties are straightforward. Be sure to try them yourself before reading the ones in the text.
The inverse of a partial order is also a partial order.
Details:
Clearly the reflexive, antisymmetric and transitive properties hold for .
If is a partial order on and is a subset of , then the restriction of to is a partial order on .
Details:
The reflexive, antisymmetric, and transitive properties given above hold for all and hence hold for all .
The following theorem characterizes relations that correspond to strict order.
Let be a set. A relation is a partial order on if and only if is transitive and irreflexive.
Details:
Suppose that is a partial order on . Recall that is defined by if and only if and . If and then and , and so . On the other hand, if then and so , a contradiction. Hence and so . Therefore is transitive. If then by definition, so is irreflexive.
Conversely, suppose that is a transitive and irreflexive relation on . Recall that is defined by if and only if or . By definition then, is reflexive: for every . Next, suppose that and . If and then by the transitive property of . But this is a contradiction by the irreflexive property, so we must have . Thus is antisymmetric. Suppose and . There are four cases:
If and then by the transitive property of .
If and then by substitution.
If and then by substitution.
If and then by the transitive property of .
In all cases we have so is transitive. Hence is a partial order on .
Monotone Sets and Functions
Partial orders form a natural setting for increasing and decreasing sets and functions. Here are the definitions:
Suppose that is a partial order on a set and that . In the following definitions, are arbitrary elements of .
is increasing if and imply .
is decreasing if and imply .
Suppose that is a set with partial order , is a set with partial order , and that . In the following definitions, are arbitrary elements of .
is increasing if and only if implies .
is decreasing if and only if implies .
is strictly increasing if and only if implies .
is strictly decreasing if and only if implies .
Recall the definition of the indicator function associated with a subset of a universal set : For , if and if .
Suppose that is a partial order on a set and that . Then
is increasing if and only if is increasing.
is decreasing if and only if is decreasing.
Details:
is increasing if and only if and implies if and only if and implies if and only if is increasing.
is decreasing if and only if and implies if and only if and implies if and only if is decreasing.
Isomorphism
Two partially ordered sets and are isomorphic if there exists a one-to-one function from onto such that if and only if , for all . The function is an isomorphism.
Suppose that the partially ordered sets and are isomorphic, and that is an isomorphism. Then and are strictly increasing.
Details:
We need to show that for , if and only if . If then by definition, . But if then since is one-to-one. This is a contradiction, so . Similarly, if then by definition, . But if then , a contradiction. Hence .
In a sense, the subset partial order is universal—every partially ordered set is isomorphic to for some collection of sets .
Suppose that is a partial order on a set . Then there exists such that is isomorphic to .
Details:
For each , let , and then let , so that . We will show that the function from onto is one-to-one, and satisfies
First, suppose that and . Then so and hence . Similarly, so and hence . Thus , so the mapping is one-to-one. Next, suppose that . If then so by the transitive property, and hence . Thus . Conversely, suppose . As before, , so and hence .
Extremal Elements
Various types of extremal elements play important roles in partially ordered sets. Here are the definitions:
Suppose that is a partial order on a set and that .
An element is the minimum element of if and only if for every .
An element is a minimal element of if and only if no satisfies .
An element is the maximum element of if and only if for every .
An element is a maximal element of if and only if no satisfies .
In general, a set can have several maximal and minimal elements (or none). On the other hand,
The minimum and maximum elements of , if they exist, are unique. They are denoted and , respectively.
Details:
Suppose that are minimum elements of . Since we have and , so by the antisymmetric property. The proof for the maximum element is analogous.
Minimal, maximal, minimum, and maximum elements of a set must belong to that set. The following definitions relate to upper and lower bounds of a set, which do not have to belong to the set.
Suppose again that is a partial order on a set and that . Then
An element is a lower bound for if and only if for every .
An element is an upper bound for if and only if for every .
The greatest lower bound or infimum of , if it exists, is the maximum of the set of lower bounds of .
The least upper bound or supremum of , if it exists, is the minimum of the set of upper bounds of .
By [20], the greatest lower bound of is unique, if it exists. It is denoted or . Similarly, the least upper bound of is unique, if it exists, and is denoted or . Note that every element of is a lower bound and an upper bound for , since the conditions in the definition hold vacuously.
The symbols and are also used for infimum and supremum, respectively, so and if they exist. In particular, for , operator notation is more commonly used, so and . Partially ordered sets for which these elements always exist are important, and have a special name.
Suppose that is a partial order on a set . Then is a lattice if and exist for every .
For the subset partial order, the inf and sup operators correspond to intersection and union, respectively:
Let be a set and consider the subset partial order on , the power set of . Let be a nonempty subset of , that is, a nonempty collection of subsets of . Then
Details:
First, for every and hence is a lower bound of . If is a lower bound of then for every and hence . Therefore is the greatest lower bound.
First, for every and hence is an upper bound of . If is an upper bound of then for every and hence . Therefore is the least upper bound.
In particular, and , so is a lattice.
Consider the division partial order on the set of positive integers and let be a nonempty subset of .
is the greatest common divisor of , usually denoted in this context.
If is infinite then does not exist. If is finite then is the least common multiple of , usually denoted in this context.
Suppose that is a set and that . An element is said to be a fixed point of if .
The following result explores a basic fixed point theorem for a partially ordered set. The theorem is important in the study of cardinality.
Suppose that is a partial order on a set with the property that exists for every . If is increasing, then has a fixed point.
Details:
Let and let . If then so . Hence is an upper bound of so . But then so . Hence . Therefore .
Note that the hypotheses of the theorem require that exists. The set is nonempty since .
If is a total order on a set with the property that every nonempty subset of has a minimum element, then is said to be well ordered by . One of the most important examples is , which is well ordered by the ordinary order . On the other hand, the well ordering principle, which is equivalent to the axiom of choice, states that every nonempty set can be well ordered.
Orders on Product Sets
Suppose that and are sets with partial orders and respectively. Define the relation on by if and only if and .
The relation is a partial order on , called, appropriately enough, the product order.
Suppose that . If has at least 2 elements, then is not a total order on .
Details:
If then and and therefore . Hence is reflexive. Suppose that and that and . Then and , and and . Hence and , so . Hence is antisymmetric. Finally, suppose that , and that and . Then and , and and . Hence and , and therefore . Hence is transitive.
Suppose that , and that are distinct. If then and and hence , a contradiction. Similarly leads to the same contradiction . Hence and are unrelated in the partial order .
The product order on . The region shaded blue is the set of points . The region shaded red is the set of points . The region shaded white is the set of points that are not comparable with .
Product order extends in a straightforward way to the Cartesian product of a finite or an infinite sequence of partially ordered spaces. For example, suppose that is a set with partial order for each , where . The product order on the product set is defined as follows: for and in the product set, if and only if for each . We can generalize this further to arbitrary product sets. Suppose that is a set for each in a nonempty (both otherwise arbitrary) index set . Recall that
To make the notation look more like a simple Cartesian product, we will write instead of for the value of a function in the product set at .
Suppose that is a set with partial order for each in a nonempty index set . Define the relation on by if and only if for each . Then is a partial order on the product set, known again as the product order.
Details:
In spite of the abstraction, the proof is perfectly straightforward. Suppose that .
for every , and hence . Thus is reflexive.
Suppose that and . Then and for each . Hence for each and so . Thus is antisymmetric
Suppose that and . Then and for each . Hence for each , so . Thus is transitive.
Note again that no assumptions are made on the index set , other than it be nonempty. In particular, no order is necessary on . The next result gives a very different type of order on a product space.
Suppose again that and are sets with partial orders and respectively. Define the relation on by if and only if either , or and .
The relation is a partial order on , called the lexicographic order or dictionary order.
If and are total orders on and , respectively, then is a total order on .
Details:
If , then and . Hence and so is reflexive. Suppose that , and that and . There are four cases, but only one can actually happen.
If and then . This is a contradiction since is irreflexive, and so this case cannot happen.
If , , and , then . This case also cannot happen.
If , , and , then . This case also cannot happen
if , , , and , then and , so .
Hence is antisymmetric. Finally, suppose that and that and . Again there are four cases:
If and then , and hence .
If , , and then and hence .
If , , and then and hence .
If , , , and , then and , and hence .
Therefore is transitive.
Suppose that and are total orders. Let . Then either or and either or . Again there are four cases
If then
If then
If and then
If and then
Hence is a total order.
The lexicographic order on . The region shaded blue is the set of points . The region shaded red is the set of points .
As with the product order, the lexicographic order can be generalized to a collection of partially ordered spaces. However, we need the index set to be totally ordered.
Suppose that is a set with partial order for each in a nonempty index set . Suppose also that is a total order on . Define the relation on the product set as follows: if and only if there exists such that if and . Then
is a partial order on , known again as the lexicographic order.
If is a total order for each , and is well ordered by , then is a total order on .
Details:
By [12], we need to show that is irreflexive and transitive. First, no satisfies since for all . Hence is irreflexive. Next, suppose that and that and . Then there exists such that if and . Similarly, there exists such that if and . Again, since is totally ordered, either or or . If , then if and . If , then if and . If , then if and . In all cases, so is transitive.
Suppose now that is a total order on for each and that is well ordered by . Let with . Let . Then by assumption, and hence has a minimum element . If then and hence . On the other hand, since and therefore, since is totally ordered, we must have either or . In the first case, and in the second case . Hence is totally ordered.
The term lexicographic comes from the way that we order words alphabetically: We look at the first letter; if these are different, we know how to order the words. If the first letters are the same, we look at the second letter; if these are different, we know how to order the words. We continue in this way until we find letters that are different, and we can order the words. In fact, the lexicographic order is sometimes referred to as the first difference order. Note also that if is a set and a total order on for , then by the well ordering principle, there exists a well ordering of , and hence there exists a lexicographic total order on the product space . As a mathematical structure, the lexicographic order is not as obscure as you might think.
is isomorphic to the lexicographic product of with , where is the ordinary order for real numbers.
Details:
Every can be uniquely expressed in the form where is the integer part and is the remainder. Thus is a one-to-one function from onto . For example, maps to , while maps to . Suppose that , where of course are the integer parts of and , respectively, and are the corresponding remainders. Then if and only if or and . Again, to illustrate with real real numbers, we can tell that just by comparing the integer parts: . We can ignore the remainders. On the other hand, to see that we need to compare the remainders: since the integer parts are the same.
Limits of Sequences of Real Numbers
Suppose that is a sequence of real numbers.
The sequence is increasing in .
Since the sequence of infimums in the last result is increasing, the limit exists in , and is called the limit inferior of the original sequence:
The sequence is decreasing in .
Since the the sequence of supremums in the last result is decreasing, the limit exists in , and is called the limit superior of the original sequence:
Note that and equality holds if and only if exists (and is the common value).
Vector Spaces of Functions
Suppose that is a nonempty set, and recall that the set of functions is a vector space, under the usual pointwise definition of addition and scalar multiplication. As noted in [9], is also a partial ordered set, under the pointwise partial order: if and only if for all . Consistent with the definitions in [21], is bounded if there exists such that for all . Now let denote the set of bounded functions , and for define
is a vector subspace of and is a norm on .
Details:
To show that is a subspace, we just have to note that it is closed under addition and scalar multiplication. That is, if are bounded, and if , then and are bounded. Next we show that satisfies the axioms of a norm. Again, let and
Clearly and if and only if for all if and only if , the zero function on .
By the usual triangle inequality on , for . Hence
That is, .
Recall that part (a) is the positive property, part (b) is the scaling property, and part (c) is the triangle inequality.
Appropriately enough, is called the supremum norm on . Vector spaces of bounded, real-valued functions, with the supremum norm are especially important in probability and random processes. We will return to this discussion again in the sections on metric spaces and measure spaces.
Computational Exercises
Let .
Sketch the covering graph corresponding to the ordinary order on .
Sketch the covering graph corresponding to the division partial order on .
Details:
The covering graph of
The covering graph of
Consider the ordinary order on the set of real numbers , and let where . Find each of the following that exist:
The set of minimal elements of
The set of maximal elements of
The set of lower bounds of
The set of upper bounds of
Details:
Does not exist
Again consider the division partial order on the set of positive integers and let . Find each of the following that exist:
The set of minimal elements of
The set of maximal elements of
The set of lower bounds of
The set of upper bounds of
.
Details:
Does not exist
Let .
Give in list form.
Describe the covering graph of
Details:
For and , there is a directed edge from to
Note that the covering graph of looks the same as the graph of , except for the labels on the vertices. This symmetry is because of the complement relationship.
Let .
Give in list form.
Describe the covering graph of
Details:
For and , there is a directed edge from to
Note again that the covering graph of looks the same as the graph of , except for the labels on the vertices. This symmetry is because the complement relationship.
Suppose that and are subsets of a universal set . Let denote the collection of the 16 subsets of that can be constructed from and using the set operations. Show that is isomorphic to the partially ordered set in the previous exercise. Use the Venn diagram app to help.
Details:
Let , , , . Our basic assumption is that and are in general position, so that are distinct and nonempty. Note also that partitions . Now, map each subset of to . This function is an isomorphism from to . That is, for and subsets of , if and only if .