Expected Value and First Order Stochastic Dominance


by odck11
Tags: means, stochastic dominance
odck11
odck11 is offline
#1
Feb4-13, 03:21 PM
P: 2
Dear All:

Given two random variables X and Y, if I have established the relationship E[X]>=E[Y], does this necessarily imply that X must have a first-order-stochastic dominance over Y?

I know that first order stochastic dominance implies that the mean value of the dominating random variable be greater than the other variable but I am trying to find out whether the reverse must hold.

Thanks in advance.
Regards.
Phys.Org News Partner Mathematics news on Phys.org
Researchers help Boston Marathon organizers plan for 2014 race
'Math detective' analyzes odds for suspicious lottery wins
Pseudo-mathematics and financial charlatanism
mathman
mathman is offline
#2
Feb4-13, 03:30 PM
Sci Advisor
P: 5,935
Quote Quote by odck11 View Post
Dear All:

Given two random variables X and Y, if I have established the relationship E[X]>=E[Y], does this necessarily imply that X must have a first-order-stochastic dominance over Y?

I know that first order stochastic dominance implies that the mean value of the dominating random variable be greater than the other variable but I am trying to find out whether the reverse must hold.

Thanks in advance.
Regards.
Not necessarily. Let X have two states 10 and 0, while Y has two states 2 and 1, both with equal probability. E(X) = 5, E(Y) = 1.5, but X does not dominate Y.
odck11
odck11 is offline
#3
Feb4-13, 03:46 PM
P: 2
Great! Thanks a lot. That's what I guessed too but just wanted to make sure. I appreciate your fast reply.


Register to reply

Related Discussions
Matter dominance Cosmology 1
Stochastic Processes, Poisson Process | Expected value of a sum of functions. Calculus & Beyond Homework 2
Order Statistics, Unbiasedness, and Expected Values Calculus & Beyond Homework 2
Incomplete Dominance and Codominance Biology 1
Expected value of a third order statistic? Calculus & Beyond Homework 1