This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: • Maximum likelihood estimator may or may not be unbiased. The bias, if present, be comes small as the sample size increases. • Maximum likelihood estimators are consis tent, and for large sample sizes their MSE is the smallest possible. • If ˆ θ is the MLE of an unknown parameter θ , and h is a onetoone function, then h ( ˆ θ ) is a MLE of h ( θ ). Estimation by the method of moments • The moments of the distribution of the sample (the population moments) depend on the unknown parameters. • One can estimate these moments from the sample, resulting in the sample moments. • Equate populations moments and the sam ple moments, and solve the resulting sys tem of equations for the unknown param eters • The solution is a moment estimator of the parameters. Sample moments : Given a sample X 1 ,...,X n , the k th sample mo ment is ˆ m k = 1 n n X i =1 X k i , k = 1 , 2 ,.......
View
Full
Document
This note was uploaded on 12/03/2010 for the course OR&IE 3500 taught by Professor Samorodnitsky during the Fall '10 term at Cornell.
 Fall '10
 SAMORODNITSKY

Click to edit the document details