# 9 - NOTES 9 Forecasting: Mean squares error prediction Give...

This preview shows pages 1–2. Sign up to view the full content.

NOTES 9 Forecasting : Mean squares error prediction Give a random vector ( Y, X ) 0 , where Y is a scalar and X is a vector. We want to get a function g ( X ) which predict Y as closely as possible, in the Mean squares error sense, i.e MSE = E [ Y - g ( X )] 2 is to be minimized. Claim : g ( X ) = E ( Y | X ) is the optimal choice. Here E ( Y | X ) is called the Mean squares error predictor. Proof of the Claim : First observe that E [( Y - c ) 2 ] = E [( Y - μ ) + ( μ - c )] 2 , μ = E ( Y ) = E ( Y - μ ) 2 + ( μ - c ) 2 + 2( μ - c ) E ( Y - μ ) = E ( Y - μ ) 2 + ( μ - c ) 2 E ( Y - μ ) 2 with equality holds if and only if c = μ = E ( Y ). That is, E [( Y - c ) 2 ] is minimized at c = E ( Y ). Recall that E [( Y - c ) 2 ] = E [( Y - E ( Y )) 2 ] + [ E ( Y ) - c ] 2 . For any function g ( X ), put c = g ( x ). Take expectation of Y given X = x . Then we have E [( Y - g ( x )) 2 | X = x ] = E [( Y - E ( Y | X = x )) 2 | X = x ] +[ E ( Y | X = x ) - g ( x )] 2 Therefore E [( Y - g ( X )) 2 | X ] = E [( Y - E ( Y

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 11/02/2011 for the course STAT 4005 taught by Professor Wu,kaho during the Spring '08 term at CUHK.

### Page1 / 4

9 - NOTES 9 Forecasting: Mean squares error prediction Give...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online