# lec4 - Lecture 4 Let us go back to the example of...

This preview shows pages 1–3. Sign up to view the full content.

Lecture 4 Let us go back to the example of exponential distribution E ( α ) from the last lecture and recall that we obtained two estimates of unknown parameter α 0 using the Frst and second moment in the method of moments. We had: 1. Estimate of α 0 using Frst moment: g ( X ) = X, m ( α ) = α g ( X ) = 1 α , ˆ α 1 = m - 1 g ) = 1 ¯ X . 2. Estimate of α using second moment: g ( X ) = X 2 , m ( α ) = α g ( X 2 ) = 2 α 2 , ˆ α 2 = m - 1 g ) = r 2 ¯ X 2 . How to decide which method is better? The asymptotic normality result states: n ( m - 1 g ) - θ 0 ) N ± 0 , Var θ 0 ( g ( X )) ( m 0 ( θ 0 )) 2 ² . It makes sense to compare two estimates by comparing their asymptotic variance. Let us compute it in both cases: 1. In the Frst case: Var α 0 ( g ( X )) ( m 0 ( α 0 )) 2 = Var α 0 ( X ) ( - 1 α 2 0 ) 2 = 1 α 2 0 ( - 1 α 2 0 ) 2 = α 2 0 . In the second case we will need to compute the fourth moment of the exponential distribution. This can be easily done by integration by parts but we will show a di±erent way to do this. The moment generating function of the distribution E ( α ) is: ϕ ( t ) = α e tX = Z 0 e tx αe - αx dx = α α - t = X k =0 t k α k , 13

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

View Full Document
LECTURE 4. 14 where in the last step we wrote the usual Taylor series. On the other hand, writing
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 10/11/2009 for the course STATISTICS 18.443 taught by Professor Dmitrypanchenko during the Spring '09 term at MIT.

### Page1 / 4

lec4 - Lecture 4 Let us go back to the example of...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online