Econometrics I 3
Econometrics I 3
Econometrics I 3
William Greene
Department of Economics
Stern School of Business
Applied Econometrics
= (Var[xi])-1 Cov[xi,yi].
1
1 1
b= X'X X'y
n n
Suggests something desirable about least squares
Second Order Conditions
= 2X'X
Does b Minimize e’e?
in1 xi21 in1 xi1 xi 2 ... in1 xi1 xiK xi21 xi1 xi 2 ... xi1 xiK
n n 2
x x x ... in1 xi 2 xiK n xi 2 xi1 xi22 ... xi 2 xiK
X'X = i 1 i 2 i1 i 1 i 2
=i 1
... ... ... ... ... ... ... ...
n n
i 1 xiK xi1 i 1 xiK xi 2 ... in1 xiK2 xiK xi1 xiK xi 2 ... xiK2
xi1
x
=in1 i 2 xi1 xi 2 ... xiK
...
xik
=in1xi xi
Positive Definite Matrix
Matrix C is positive definite if a'Ca is > 0
for any x.
Generally hard to check. Requires a look at
characteristic roots (later in the course).
For some matrices, it is easy to verify. X'X is
one of these.
k=1 k 0
K 2
a'X'Xa = (a'X )( X'a) = ( X'a)'( X'a) = v'v = v
Could v = 0?
Conclusion: b = ( X'X)-1 X'y does indeed minimize e'e.
Algebraic Results - 1
n
i'y = i=1
y i ny
i'y i'Xb + i'e = i'Xb
implies (after dividing by n)
y x b (the regression line passes through the means)
These do not apply if the model has no constant term.
Least Squares Algebra
Least Squares
Residuals
Least Squares Residuals
Least Squares Algebra-3