slides_Ch2_W[1]

# Slides_Ch2_W[1]

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: β1 = 1 Ex 2 2 =4 2 Ey 2 µy =4 =4 Melissa Tartari (Yale) (Ex ) Ex 2 Ey Ex 2 Exy Ex 2 2 E xExy (Ex )2 E xEy (Ex )2 Ex 2 Ex 3 5 E xy E xEy Ex Ex 2 (Ex )2 Exy E xEy Ex 2 (Ex )2 Ex 1 3 5 E [(y µy )(x µx )] µx E [(x µx )2 ] E [(y µy )(x µx )] E [(x µx )2 ] Econometrics Ey Eyx 3 5 (4) 24 / 93 The Method of Moments Approach IV Observe that (4) express the uknown parameters ( βo , β1 ) as functions n o of the unknown population objects µy , µx , var (x ) , cov (x , y ) . Have we then made any progress? Melissa Tartari (Yale) Econometrics 25 / 93 The Method of Moments Approach IV Observe that (4) express the uknown parameters ( βo , β1 ) as functions n o of the unknown population objects µy , µx , var (x ) , cov (x , y ) . Have we then made any progress? It is true that the expressions in (4) do not provide us with estimators of ( βo , β1 ); however, they suggest estimators. How? Melissa Tartari (Yale) Econometrics 25 / 93 The Method of Moments Approach IV Observe that (4) express the uknown parameters ( βo , β1 ) as functions n o of the unknown population objects µy , µx , var (x ) , cov (x , y ) . Have we then made any progress? It is true that the expressions in (4) do not provide us with estimators of ( βo , β1 ); however, they suggest estimators. How? The idea of the MoM approach is to replace the unknown population n o objects µy , µx , var (x ) , cov (x , y ) in (4) with their sample counterparts (or analogues) by exploting the LLN which tells us that: 1n 1n 1n p p xi ! µx , ∑ yi ! µy , ∑ (xi n i∑ n i =1 n i =1 =1 and Melissa Tartari (Yale) 1n (yi n i∑ =1 y ) (xi Econometrics p x )2 ! Var (x ) ¯ p x ) ! cov (x , y ) . ¯ 25 / 93 The Method of Moments Approach V Thus, using (4) we de…ne MoM estimators: β1 = βo cov (x , y ) Var (x ) = µy Melissa Tartari (Yale) β1 µx MoM Approach ) MoM Approach ) ¯ ∑n (yi y ) (xi x ) ˆ MM (2.19) β 1 = i =1 n 2 ¯ ∑i =1 (xi x ) ˆ MM βo = y ¯ Econometrics ˆ MM ¯ β1 x (2.17) 26 / 93 The Method of Moments Approach VI More generally, the LLN allows us to rewrite the system (2.1...
View Full Document

## This note was uploaded on 02/13/2014 for the course ECON 350 taught by Professor Donaldbrown during the Fall '10 term at Yale.

Ask a homework question - tutors are online