STAT 111 Chapter 7 Functions of Random Variables Frequently, in statistics, one encounters the need to derive the probability distribution of a function of one or more random variables. Suppose first, that X is a discrete random variable with probability distribution fx(x) and suppose further that Y=g(X) defines a one-to-one transformation between the values of X and Y(this condition implies that there exists an inverse( function g-1 such that g-1(y)=X, which can be obtained by solving the original defining relation for x. When considering random variables, we usually write g-1 (Y)=X. For one-toone transformation we have the following theorem. Note that a function Y=g(X) is a one-to-one if each value assumed by y only a single value of X is obtained. 2 Theorem Suppose that X is a discrete random variable with probability distribution fx(x). Let Y=g(X) define a one-to-one transformation between the values of X and Y. Then the probability distribution of Y is ƒy ( y ) = ƒx [g-1 (y )] 3 Example Let X be a random variable with the following distribution x f(x) 2 1/8 4 1/8 6 1/4 8 1/2 Find the probability distribution of the random variable Y=3X, As X takes the values 2,4,6,8, y takes the values 6,12,18,24, which implies that Y ,Defines a one -to-one transformation . From Y = 3 X g-1 (y ) =X = Y / 3 fY ( 6 ) = fx( 6 / 3 ) = fx(2) = 1 / 8 Similarly fy ( 12 ) = fx(12/3 ) = fx(4)=1/8 and so on . The probability distribution of Y is y f(y) 6 1/8 12 1/8 18 1/4 24 1/2 4 Example Let X be a geometric random variable with probability distribution ƒx(x) = 3 ( 1 ) x-1 x = 1 , 2 ,3 , ….. 4 4 Find the probability distribution of the random variable Y=X2. Y=X2 is not generally a 1-1 transformation, however if the range of X consists only a non-negative numbers, then the transformation is 1-1 (in this example, as x takes the values 1,2,3,..., y takes 1,4,9,...) From Y= X2 we have X= √ y = g-1 (y) => ƒx ( √ y 3 1 ) = 4 4 y 1 y= 1 , 4 , 9 , …. 5 Example Let X has the binomial random variable with probability distribution x ƒ(x( = 3 x 2 3 3-x 1 3 x =0,1,2,3 Find the probability distribution of the random variable Y=X2 As x takes 0,1,2,3 y takes 0,1,4,9 6 To derive the probability distribution of a function of more than one random variables, that is, given n random variables X1,X2,...,X n and need to find the distribution of Y=g(X1,X2,...,Xn). One straightforward method is by using the following theorem. Theorem Let X1,X2,…,Xn be n random variables defined on (S,P). Let Y=g(X1,X2,...,Xn )be a function of X1X2,...,Xn and let y є R(Y(. Then f(y) = P(Y = y)= ∑ f ( x1 , x2 …,xn) x1 , x2 ,… , xn Where Ay = { (x1 , x2 , … , xn ) є R ( X ) : g ( x1 , x2 ,… , xn ) = y } 7 Example Let the joint distribution of X and Y be given as xy f(x,y) = — 36 x = 1,2,3 and Find 1.The probability mass function of Z = y = 1,2,3 X Y Solution: The first step is to determine the range of the random , variable involved, i.e as x,y X takes the values x=l,2,3, y= 1,2,3 , then Z = Y takes the values 1, 1 , 1 , 2 ,1 , 2 , 3 , 3 , 1 2 3 3 2 8 Then P( Z = 1 ) = P ( X = 1 , Y = 3 ) = 3 = 1 3 36 12 P ( Z = 1 ) = P ( X = 1 ,Y = 1 ) + P ( X = 2 , Y = 2 ) +P ( X = 3 ,Y = 3 ) = 14 36 = 7 18 and so on Z 1/3 1/2 2/3 1 3/2 2 3 Sum f(z) 1/12 1/18 1/6 7/18 1/6 1/18 1/12 1 9 2- The probability mass function of M=XY m 1 2 3 f(m) 1/36 4/36 6/36 4 6 9 4/36 12/36 9/36 Sum 1 3- The probability mass function of T=X+Y t 2 3 4 5 6 Sum f(t) 1/36 4/36 10/36 12/36 9/36 1 10 4- The probability mass function of U=X – Y u -2 -1 0 1 2 Sum f(u) 3/36 8/36 14/36 8/36 3/36 1 11 Sum of Independent Random Variables *. Let X and Y be independent random variables where X~bin(n,p) and Y~bin(m,p), Then the distribution of X+Y is bin(n+m,p), *. Let X and Y be independent random variables where X~P(λ) and Y~P(ϒ(, Then the distribution of X+Y is P(λ+ϒ). 12 Example 1 Let X and Y be independent random variables where X~bin(2,0.6) and Y~bin(5,0.6), let Z=X+Y 1.P(Z>3) 2.P(3<Z<5) 3.fz(2)=P(Z=2) 4.P(Z<5) 13 Example Let X and Y be independent random variables where X~Poisson(3) and Y~Poisson(2), let W=X+Y Find: P(W<2) P(3<W<5) fw(2)=P(W=2) 14
© Copyright 2024 Paperzz