ECE313.Lecture40

ECE313.Lecture40 - ECE 313 Probability with Engineering...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Mean-Square Estimation Professor Dilip V. Sarwate Department of Electrical and Computer Engineering © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign. All Rights Reserved ECE 313 Probability with Engineering Applications
Background image of page 1

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

View Full DocumentRight Arrow Icon
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved Slide 2 of 40 Y is a RV with known pmf or pdf Problem: Predict (or estimate ) what value of Y will be observed on the next trial What value should we predict? What is a good prediction? We need to specify some criterion that determines what is a good/reasonable estimate Else any estimate is just as good as any other estimate Predicting the value of Y
Background image of page 2
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved Slide 3 of 40 ê denotes our estimate of the value of Y ê is a number that we get to choose Minimum-probability-of-error criterion: choose α so as to minimize P{ Y ≠ ê} If Y is a discrete random variable, then P{ Y ≠ ê} = 1 – P{ Y = ê} = 1 – p Y (ê) Choose ê to be the location of the maximum of the pmf p Y (u) Our estimate is wrong with probability 1 – p Y (ê) Minimize probability of error – I
Background image of page 3

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

View Full DocumentRight Arrow Icon
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved Slide 4 of 40 If Y is a continuous random variable, then P{ Y ≠ ê} = 1 – P{ Y = ê} = 1 no matter what we number we choose as ê Alternative: choose ê to minimize P{| Y –ê| > ε } for some suitable (small) choice of ε P{| Y –ê| > ε } = 1 – P{| Y –ê| ≤ ε } and so we want to maximize P{| Y –ê| ≤ ε } P{| Y –ê| ≤ ε } = P{ê – ε Y ≤ ê + ε } = F Y (ê + ε ) – F Y (ê – ε ) Minimize probability of error – II
Background image of page 4
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved Slide 5 of 40 P{| Y –ê| ≤ ε } = F Y (ê + ε ) – F Y (ê – ε ) has derivative f Y (ê + ε ) – f Y (ê – ε ) = 0 if ê is chosen such that f Y (ê + ε ) = f Y (ê – ε ) Graphically, find a horizontal “chord” of length 2 ε under the “peak” of the pdf: the midpoint of the chord is ê Minimize probability of error – III ê 2 ε
Background image of page 5

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

View Full DocumentRight Arrow Icon
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved Slide 6 of 40 In the limit as ε 0, ê approaches the location of the maximum value of the pdf For both continuous and discrete RVs, we get the location of the maximum of the pdf or the pmf The location of the maximum of the pdf or the pmf is called the mode of the pdf/pmf It is the value of Y that has the “maximum probability” of occuring Mode = most fashionable or most frequent What’s apple pie à la mode?
Background image of page 6
ECE 313 - Lecture 40 © 2000 Dilip V. Sarwate, University of Illinois at Urbana-Champaign, All Rights Reserved
Background image of page 7

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

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 42

ECE313.Lecture40 - ECE 313 Probability with Engineering...

This preview shows document pages 1 - 8. Sign up to view the full document.

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