Brief introduction to stochastic ordering
In probability theory and statistics, a stochastic order quantifies the concept of one random variable being "bigger" than another. These are usually partial orders, so that one random variable
Contents
- Brief introduction to stochastic ordering
- Usual stochastic order
- Characterizations
- Other properties
- Stochastic dominance
- Multivariate stochastic order
- Hazard rate order
- Likelihood ratio order
- Variability orders
- Convex order
- Laplace transform order
- Realizable monotonicity
- References
Usual stochastic order
A real random variable
where
Characterizations
The following rules describe cases when one random variable is stochastically less than or equal to another. Strict version of some of these rules also exist.
-
A ⪯ B if and only if for all non-decreasing functionsu ,E [ u ( A ) ] ≤ E [ u ( B ) ] . - If
u is non-decreasing andA ⪯ B thenu ( A ) ⪯ u ( B ) - If
u : R n → R is an increasing function andA i B i A i ⪯ B i i , thenu ( A 1 , … , A n ) ⪯ u ( B 1 , … , B n ) and in particular∑ i = 1 n A i ⪯ ∑ i = 1 n B i i th order statistics satisfyA ( i ) ⪯ B ( i ) - If two sequences of random variables
A i B i A i ⪯ B i i each converge in distribution, then their limits satisfyA ⪯ B . - If
A ,B andC are random variables such that∑ c Pr ( C = c ) = 1 andPr ( A > u | C = c ) ≤ Pr ( B > u | C = c ) for allu andc such thatPr ( C = c ) > 0 , thenA ⪯ B .
Other properties
If
Stochastic dominance
Stochastic dominance is a stochastic ordering used in decision theory. Several "orders" of stochastic dominance are defined.
Multivariate stochastic order
An
Other types of multivariate stochastic orders exist. For instance the upper and lower orthant order which are similar to the usual one-dimensional stochastic order.
and
All three order types also have integral representations, that is for a particular order
Hazard rate order
The hazard rate of a non-negative random variable
Given two non-negative variables
or equivalently if
Likelihood ratio order
Let
Variability orders
If two variables have the same mean, they can still be compared by how "spread out" their distributions are. This is captured to a limited extent by the variance, but more fully by a range of stochastic orders.
Convex order
Convex order is a special kind of variability order. Under the convex ordering,
Laplace transform order
Laplace transform order compares both size and variability of two random variables. Similar to convex order, Laplace transform order is established by comparing the expectation of a function of the random variable where the function is from a special class:
Realizable monotonicity
Considering a family of probability distributions