11 - 0.8 Estimators 0.7 0.6 0.5 0.4 Maximum a Posteriori...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
2 4 6 8 10 12 14 16 18 20 0.4 0.5 0.6 0.7 0.8 n Estimators Maximum a Posteriori Estimates Conditional Expectation Estimates 2 4 6 8 10 12 14 16 18 20 0 0.01 0.02 0.03 0.04 n Mean Squared Errors Figure 8.4: Asymptotic behavior of the MAP and LMS estimators, and the corresponding conditional mean squared errors, for Fxed x =0 . 5, and n →∞ in Problem 8.12. We have E [ X ]= E [Θ] + E [ W E [Θ] 2 X = σ 2 Θ + σ 2 W , cov(Θ ,X )= E ±( Θ E [Θ] )( X E [ X ] = E ±( Θ E [Θ] ) 2 ² = σ 2 Θ , where the last relation follows from the independence of Θ and W . Using the formulas for the mean and variance of the uniform PDF, we have E [Θ] = 7 2 Θ =3 , E [ W ]=0 2 W =1 / 3 . Thus, the linear LMS estimator is ˆ Θ=7+ 3 3+1 / 3 ( X 7 ) , or ˆ 9 10 ( X 7 ) . The mean squared error is (1 ρ 2 ) σ 2 Θ .W ehav e ρ 2 = cov(Θ ) σ Θ σ X 2 = σ 2 Θ σ Θ σ X 2 = σ 2 Θ σ 2 X = 3 / 3 = 9 10 . 101
Background image of page 1

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

View Full DocumentRight Arrow Icon
Hence the mean squared error is (1 ρ 2 ) σ 2 Θ = 1 9 10 · 3= 3 10 . Solution to Problem 8.15. The conditional mean squared error of the MAP esti- mator ˆ Θ= X is E ± ( ˆ Θ Θ) 2 | X = x ² = E ± ˆ Θ 2 2 ˆ ΘΘ+Θ 2 | X = x ² = x 2 2 x E | X = x ]+ E 2 | X = x ] = x 2 2 x 101 x 100 ³ i = x 1 i + 100 ³ i = x i 100 ³ i = x 1 i . The conditional mean squared error of the LMS estimator ˆ 101 X 100 ³ i = X 1 i . is E [( ˆ Θ Θ) 2 | X = x ]= E [ ˆ Θ 2 2 ˆ 2 | X = x ] = 101 x 2 100 ³ i = x 1 i 2 101 x 100 ³ i = x 1 i E | X = x E 2 | X = x ] = (101 x ) 2 100 ³ i = x 1 i 2 + 100 ³ i = x i 100 ³ i = x 1 i . To obtain the linear LMS estimator, we compute the expectation and variance of X .W ehav e E [ X E ± E [ X | Θ] ² = E h Θ+1 2 i = (101 / 2) + 1 2 =25 . 75 , and var( X )= E [ X 2 ] ( E [ X ] ) 2 = 1 100 100 ³ x =1 x 2 ´ 100 ³ θ = x 1 θ ! (25 . 75) 2 = 490 . 19 . 102
Background image of page 2
Thecovar ianceo fΘand X is cov(Θ ,X )= E ± ( X E [ X ])(Θ E [Θ]) ² = 100 ³ θ =1 1 100 θ ³ x =1 1 θ ( x 25 . 75)( θ 50) = 416 . 63 . Applying the linear LMS formula yields ˆ Θ= E [Θ] + cov(Θ ) var( X ) ( X E [ X ] ) =50+ 416 . 63 490 . 19 ( X 25 . 75) = 0 . 85 X +28 . 11 . The mean squared error of the linear LMS estimator is E ± ( ˆ Θ Θ) 2 | X = x ² = E ± ˆ Θ 2 2 ˆ ΘΘ+Θ 2 | X = x ² = ˆ Θ 2 2 ˆ Θ E | X = x ]+ E 2 | X = x ] =(0 . 85 x . 11) 2 2(0 . 85 x . 11) 101 x 100 i = x 1 i + 100 i = x i 100 i = x 1 i . Figure 8.5 plots the conditional mean squared error of the MAP, LMS, and linear LMS estimators, as a function of x . Note that the conditional mean squared error is lowest for the LMS estimator, but that the linear LMS estimator comes very close. 0
Background image of page 3

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

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/11/2011 for the course MATH 170 taught by Professor Staff during the Spring '08 term at UCLA.

Page1 / 10

11 - 0.8 Estimators 0.7 0.6 0.5 0.4 Maximum a Posteriori...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online