Econometrics-I-13

# N(2 linearly regress y on these k predictions p this

This preview shows pages 27–34. Sign up to view the full content.

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

View Full Document

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

View Full Document

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

View Full Document

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

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

Unformatted text preview: n (2) Linearly regress y on these K predictions. p This is two stage least squares &#152;&#152;&#152;™™™™ ™ 26/61 Part 13: Endogeneity Algebraic Equivalence p Two stage least squares is equivalent to n (1) each variable in X that is also in Z is replaced by itself. n (2) Variables in X that are not in Z are replaced by predictions of that X with all the variables in Z that are not in X . &#152;&#152;&#152;™™™™ ™ 27/61 Part 13: Endogeneity 2SLS Algebra &#152;&#152;&#152;&#152;™™™ ™ 28/61 1 1 ˆ ˆ ˆ ˆ ( ) But, = ( ) and ( ) is idempotent. ˆ ˆ ( )( ) ( ) so ˆ ˆ ( ) = a real IV estimator by the definition. ˆ Note, plim( /n) = -- = = =-1 2SLS-1 Z Z Z Z Z 2SLS X Z(Z'Z) Z'X b X'X X'y Z(Z'Z) Z'X I - M X I - M X'X = X' I - M I - M X = X' I - M X b X'X X'y X' ε-1 ˆ since columns of are linear combinations of the columns of , all of which are uncorrelated with ( ) ] ( ) 2SLS Z Z X Z b X' I - M X X' I - M y ε. =[ Part 13: Endogeneity Asymptotic Covariance Matrix for 2SLS &#152;&#152;&#152;&#152;™™™ ™ 29/61 2 1-1 IV IV 2 1-1 2SLS 2SLS General Result for Instrumental Variable Estimation E[( )( ) '| ] ( ) ' ( ) ˆ Specialize for 2SLS, using = ( ) ˆ ˆ ˆ ˆ E[( )( ) '| ] ( ) ' ( ) ---- = σ-- = σ Z b b X, Z Z'X Z Z X'Z Z X = I - M X b b X, Z X'X X X X'X β β β β 2 1-1 2 1 ˆ ˆ ˆ ˆ ˆ ˆ ( ) ' ( ) ˆ ˆ ( )-- = σ = σ X'X X X X'X X'X Part 13: Endogeneity 2SLS Has Larger Variance than LS &#152;&#152;&#152;&#152;™™™ ™ 30/61 2-1 2-1 A comparison to OLS ˆ ˆ Asy.Var[2SLS]= ( ' ) Neglecting the inconsistency, Asy.Var[LS] = ( ' ) (This is the variance of LS around its mean, not ) Asy.Var[2SLS] Asy.Var[LS] in the matrix sense. Com σ σ ≥ X X X X β-1-1 2 2 2 Z Z pare inverses: ˆ ˆ { Asy.Var[LS]} - { Asy.Var[2SLS]} (1 / )[ ' ' ] (1 / )[ ' '( ) ]=(1 / )[ ' ] This matrix is nonnegative definite. (Not positive definite as it might have some rows and columns = σ = σ- σ X X - X X X X - X I M X X M X which are zero.) Implication for "precision" of 2SLS. The problem of "Weak Instruments" Part 13: Endogeneity Estimating σ2 &#152;&#152;&#152;&#152;™™™ ™ 31/61 2 2 n 1 i 1 i n Estimating the asymptotic covariance matrix - a caution about estimating . ˆ Since the regression is computed by regressing y on , one might use ˆ (y ) ˆ This is i = σ σ = Σ- 2sls x x'b 2 n 1 i 1 i n nconsistent. Use (y ) ˆ (Degrees of freedom correction is optional. Conventional, but not necessary.) = σ = Σ- 2sls x'b Part 13: Endogeneity Cornwell and Rupert Data Cornwell and Rupert Returns to Schooling Data, 595 Individuals, 7 Years...
View Full Document

{[ snackBarMessage ]}

### Page27 / 62

n(2 Linearly regress y on these K predictions p This is two...

This preview shows document pages 27 - 34. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online