5102-Lecture-07

5102-Lecture-07 - I Let ˆ η be the mle of η . I Let g be...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture 7 Stat 5102-004 2 February 2011
Background image of page 1

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

View Full DocumentRight Arrow Icon
Maximum Likelihood Estimators Maximum likelihood estimators are defined by maximizing the likelihood. For each y ∈ Y , let ˆ θ ( y ) = arg max θ lik( θ | y ) . ˆ θ : Y → ϑ y 7→ arg max θ lik( θ | y ) Since the logarithm is an increasing function, arg max θ ( θ | y ) = arg max θ lik( θ | y ) . The same arguments maximize both the likelihood and log likelihood. STAT 5102 (Theory of Statistics) Lecture 7 1 / 6
Background image of page 2
Poisson Example λ = E Y λ lik ( λ |y 29 y λ log lik y STAT 5102 (Theory of Statistics) Lecture 7 2 / 6
Background image of page 3

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

View Full DocumentRight Arrow Icon
Poisson Example θ = log E Y θ lik ( θ |y 29 log y θ log lik log y STAT 5102 (Theory of Statistics) Lecture 7 3 / 6
Background image of page 4
Invariance of the mle θ = log λ λ log lik ( λ |y 29 y θ θ log y STAT 5102 (Theory of Statistics) Lecture 7 4 / 6
Background image of page 5

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

View Full DocumentRight Arrow Icon
Invariance of the mle The maximum likelihood estimator is invariant to reparameterizations.
Background image of page 6
Background image of page 7
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: I Let ˆ η be the mle of η . I Let g be a bijection, and let θ = g ( η ) . I The mle of θ is ˆ θ = g (ˆ η ) . 1. Densities determine the values of a likelihood function. 2. Relabeling distributions has no effect on their densities. STAT 5102 (Theory of Statistics) Lecture 7 5 / 6 Y 1 , . . . , Y 12 iid ∼ Bern ( p ) ¯ y = 2 3 p lik ( p|y 29 y 0.0 0.2 0.4 0.6 0.8 1.0 0.00 0.25 ϖ ϖ |y y 1-y ϖ = p 1-p 1 2 3 4 5 μ μ n y μ = n p 2 4 6 8 10 12 θ θ logit y θ = logit p-4-2 2 4 STAT 5102 (Theory of Statistics) Lecture 7 6 / 6...
View Full Document

This note was uploaded on 02/24/2011 for the course STAT 5102 taught by Professor Staff during the Spring '03 term at Minnesota.

Page1 / 7

5102-Lecture-07 - I Let ˆ η be the mle of η . I Let g be...

This preview shows document pages 1 - 7. Sign up to view the full document.

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