This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Partitioning : = E[  ] + Residuals (sample) y e Partitio x y y y X x b β ε = conditional mean + disturbance ning : = + y y Xb X' e = projection + residual ( Note : Projection 'into the column space of ) Part 3: Least Squares Algebra Algebraic Results  2 p A “residual maker” M = ( I  X ( X’X )1 X’ ) p e = y  Xb = y  X ( X’X )1 X’y = My p My = The residuals that result when y is regressed on X p MX = 0 (This result is fundamental!) How do we interpret this result in terms of residuals? When a column of X is regressed on X, we get a perfect fit and zero residuals. p (Therefore) My = MXb + Me = Me = e (You should be able to prove this. p y = Py + My, P = X ( X’X )1 X’ = (I  M). PM = MP = 0. p Py is the projection of y into the column space of X . ˜˜˜™ ™ 18/26 Part 3: Least Squares Algebra The M Matrix p M = I X(X’X)1X’ is an nxn matrix p M is symmetric – M = M ’ p M is idempotent – M * M = M (just multiply it out) p M is singular – M1 does not exist. (We will prove this later as a side result in another derivation.) ˜˜˜™ ™ 19/26 Part 3: Least Squares Algebra Results when X Contains a Constant Term p X = [ 1 , x 2,…, x K] p The first column of X is a column of ones p Since X’e = , x1’e = 0 – the residuals sum to zero. ˜˜˜™ ™ 20/26 = = = = = ′ = ∑ + n i i=1 Define [1,1,...,1] ' a column of n ones = y ny implies (after dividing by n) y (the regression line passes through the means) These do not apply if the model has no y Xb e i i'y i'y i'Xb + i'e = i'Xb x b constant term. Part 3: Least Squares Algebra Least Squares Algebra ˜˜˜™ ™ 21/26 Part 3: Least Squares Algebra Least Squares ˜˜˜™ ™ 22/26 Part 3: Least Squares Algebra Residuals ˜˜˜ ™ 23/26 Part 3: Least Squares Algebra Least Squares Residuals ˜˜˜ ˜™ 24/26 Part 3: Least Squares Algebra Least Squares Algebra3 M is nxn potentially huge ˜˜˜ ˜™ 25/26 I X ′ X ′ X X M ′ X e Part 3: Least Squares Algebra Least Squares Algebra4 MX = ˜˜˜ ˜ 26/26...
View
Full Document
 Fall '10
 H.Bierens
 Econometrics, Yi, Linear least squares, Σi, Stern School of Business, Squares Algebra, Professor William Greene

Click to edit the document details