This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Stat 443  Forecasting Homework Assignment 1 Solutions Fall 2010 1. By the CauchySchwarz inequality, for any two vectors, say x and y in an inner product space,  < x,y >   x  y  where < , > is an inner product and  x  = < x,x > 1 / 2 . Furthermore, it states that equality holds if and only if x and y are linearly dependent. Since the expectation of a product of random variables is an inner product, we de ne < X,Y > = E ( XY ) for some random variables X and Y . Consider the random variables U E ( U ) and V E ( V ) . We have that < U E ( U ) ,V E ( V ) > = E ( ( U E ( U ) )( V E ( V ) ) ) = Cov ( U,V ) ,  U E ( U )  = < U E ( U ) ,U E ( U ) > 1 / 2 = E (( U E ( U )) 2 ) = V ar ( U ) , and  V E ( V )  = V ar ( V ) Therefore, by the CauchySchwarz inequality,  Cov ( U,V )  = V ar ( U ) V ar ( V ) if and only if U E ( U ) and V E ( V ) are linearly dependent. Rearranging, we have that  ( U,V )  =  Cov ( U,V )  V ar ( U ) V ar ( V ) = 1 i U E ( U ) and V E ( V ) are linearly dependent, that is, i V E ( V ) = a ( U E ( U )) = V = aU + ( E ( V ) aE ( U )) for some a R . This implies that there exists some linear relationship between U and V i  ( U,V )  = 1 . Note: Instead of using the CauchySchwarz inequality, you can use the fact that the angle, between two vectors, say x and y is given by cos( ) = < x, y >  x  y  Alternative Proof: " = ": Suppose that there exists a linear relationship between U and V , i.e. V = a + bU , 1 where a,b R and b = 0 . Then, by the de nition of correlation, ( U,V ) = Cov ( U,V ) V ar ( U ) V ar ( V ) = Cov ( U,a + bU ) V ar...
View
Full
Document
This note was uploaded on 12/12/2010 for the course ACTSC 445 taught by Professor Christianelemieux during the Fall '09 term at Waterloo.
 Fall '09
 ChristianeLemieux

Click to edit the document details