Econometrics-I-12

Econometrics-I-12 - Applied Econometrics William Greene...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Applied Econometrics William Greene Department of Economics Stern School of Business
Background image of page 1

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

View Full DocumentRight Arrow Icon
Applied Econometrics 12. Asymptotics for the Least Squares Estimator in the Classical Regression Model
Background image of page 2
Setting The least squares estimator is         ( X X ) -1 X y   =  ( X X ) -1 Σ i x i y i                            =   β    + ( X X ) -1 Σ i x i ε i   So, it is a constant vector plus a sum of random  variables. Our ‘finite sample’ results  established the behavior of the sum according  to the rules of statistics.  The question for the  present is how does this sum of random  variables behave in large samples? 
Background image of page 3

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

View Full DocumentRight Arrow Icon
Well Behaved Regressors A crucial assumption:   Convergence of the  moment matrix  X X /n to a positive definite  matrix of finite elements,  Q  What kind of data will satisfy this assumption?   What won’t?         Does stochastic vs. nonstochastic matter? Various conditions for “well behaved  X
Background image of page 4
Probability Limit - = - = = + × ε ÷ ÷ × ε × ε ÷ ÷ β β29( β29 = 1 n i i i 1 1 n i i i i i 1 i We use convergence in mean square.  Adequate for  almost all problems, not adequate for some time  series problems. 1 1 n n 1 1 1 ( ' ' n n n b X'X x b - b - X'X x x - = - - =  ÷ ÷     ε ε ÷  ÷ ÷    ∑ ∑ = β29( β29 β29( β29 1 n 1 1 1 n i i j j 2 i 1 1 n 1 1 1                    ' n n n In E[( '| ] in the double sum, terms with unequal subscripts have expectation zero. E[( ' | n j=1 X'X X'X x x X'X b - b - X b - b - - - = - - -    ε ÷  ÷ ÷    σ σ    = = ÷  ÷ ÷ ÷    = 1 1 n 2 i j i 2 i 1 1 1 1 2 2 1 1 1 'E[ | ] n n n 1 1 1 1               n n n n n n X X'X x x X X'X X'X X'X X'X X'X
Background image of page 5

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

View Full DocumentRight Arrow Icon
Mean Square Convergence E[ b | X ]= β  for any X. Var[ b | X ] 0 for any specific  X b  converges in mean square to  β
Background image of page 6
= ÷ ε ε i i 1 What must be assumed to get plim ? n (1)   = a random vector with finite means and variance and identical distributions. (2)   = a random variable with a constant distribution with finite mean X' 0 x = ε ε ε i i i i i i n i i 1  and variance and E[ ]=0 (3)   and   statistically independent. Then,   =   = an observation in a random sample, with constant variance matrix and mean vector 0. 1
Background image of page 7

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

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/23/2011 for the course ECON B30.3351 taught by Professor Professorw.greene during the Spring '10 term at NYU.

Page1 / 39

Econometrics-I-12 - Applied Econometrics William Greene...

This preview shows document pages 1 - 8. Sign up to view the full document.

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