Part 4: Partial Regression and Correlation -1/22 Econometrics I Professor William Greene Stern School of Business Department of Economics
Feb 25, 2016
Part 4: Partial Regression and Correlation4-1/22
Econometrics I Professor William GreeneStern School of Business
Department of Economics
Part 4: Partial Regression and Correlation4-2/22
Econometrics I
Part 4 – Partial Regression and Correlation
Part 4: Partial Regression and Correlation4-3/22
Frisch-Waugh (1933) Theorem
Context: Model contains two sets of variables: X = [ (1,time) : (other variables)] = [X1 X2] Regression model: y = X11 + X22 + (population) = X1b1 + X2b2 + e (sample)Problem: Algebraic expression for the second set
of least squares coefficients, b2
Part 4: Partial Regression and Correlation4-4/22
Partitioned SolutionDirect manipulation of normal equations produces
, ] so and
( ) =
1 1 1 2 11 2
2 1 2 2 2
1 1 1 2 1 1
2 1 2 2 2 2
1 1 1 1 2 2 1
2 1 1 2 2 2 2 2 2 2 2 2 1 1
X X b X yX X X X X y
X = [X X X X = X y =X X X X X y
X X X X b X y(X X)b = =
X X X X b X yX X b X X b X yX X b X X b X y ==> X X b X y - X X b = 2 1 1X (y - X b )
Part 4: Partial Regression and Correlation4-5/22
Partitioned Solution
Direct manipulation of normal equations produces b2 = (X2X2)-1X2(y - X1b1) What is this? Regression of (y - X1b1) on X2
If we knew b1, this is the solution for b2.Important result (perhaps not fundamental). Note
the result if X2X1 = 0.
Part 4: Partial Regression and Correlation4-6/22
Partitioned Inverse Use of the partitioned inverse result
produces a fundamental result: What is the southeast element in the inverse of the moment matrix?
1 1 1 2
2 1 2 2
-1X 'X X 'XX 'X X 'X
Part 4: Partial Regression and Correlation4-7/22
Partitioned Inverse The algebraic result is: [ ]-1
(2,2) = {[X2’X2] - X2’X1(X1’X1)-1X1’X2}-1
= [X2’(I - X1(X1’X1)-1X1’)X2]-1
= [X2’M1X2]-1 Note the appearance of an “M” matrix.
How do we interpret this result?
Part 4: Partial Regression and Correlation4-8/22
Frisch-Waugh ResultContinuing the algebraic manipulation:
b2 = [X2’M1X2]-1[X2’M1y].
This is Frisch and Waugh’s famous result - the “double residual regression.”
How do we interpret this? A regression of residuals on residuals.
Regrido Y em X1 – pego o resíduo que sobra e regrido esta parte no que sobra de X2 regredido em X1 (resíduo).
Tudo de Y que não foi explicado por X1 mas pode ser explicado por X2.
Tudo de X2 que não está sendo explicado por X1
Part 4: Partial Regression and Correlation4-9/22
Part 4: Partial Regression and Correlation4-10/22
Important Implications M1 is idempotent. (This is a very useful result.)
(i.e., X2M1’M1y = X2M1y)
(Orthogonal regression) Suppose X1 and X2 are orthogonal; X1X2 = 0. What is M1X2?
Part 4: Partial Regression and Correlation4-11/22
Applying Frisch-Waugh
Using gasoline data from Notes 3.X = [1, year, PG, Y], y = G as before.
Full least squares regression of y on X.
Part 4: Partial Regression and Correlation4-12/22
Detrending the Variables - Pg
Part 4: Partial Regression and Correlation4-13/22
Regression of Detrended G on Detrended Pg and Detrended Y
Part 4: Partial Regression and Correlation4-14/22
Partial RegressionImportant terms in this context: Partialing out the effect of X1. Netting out the effect …
“Partial regression coefficients.” To continue belaboring the point: Note the interpretation of partial
regression as “net of the effect of …”
Now, follow this through for the case in which X1 is just a constant term, column of ones.
What are the residuals in a regression on a constant. What is M1? Note that this produces the result that we can do linear regression on data
in mean deviation form.As inclinações numa regressão múltipla que contém o termo
constante são obtidas por transformar os dados como desvios em relação à média e regredir a variável y na forma de desvio em todas variáveis explicativas também como desvios em relação às suas médias.
'Partial regression coefficients' are the same as 'multiple regression coefficients.' It follows from the Frisch-Waugh theorem.
Part 4: Partial Regression and Correlation4-15/22
Partial CorrelationWorking definition. Correlation between sets of residuals. Some results on computation: Based on the M matrices. Some important considerations: Partial correlations and coefficients can have signs and magnitudes that differ greatly from gross correlations and
simple regression coefficients.
Compare the simple (gross) correlation of G and PG with the partial correlation, net of the time effect.
CALC;list;Cor(g,pg)$ Result = .7696572
CALC;list;cor(gstar,pgstar)$ Result = -.6589938
G
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
.5080 90 100 110 12070
PG
Part 4: Partial Regression and Correlation4-16/22
Partial Correlation
Part 4: Partial Regression and Correlation4-17/22
THE Application of Frisch-Waugh The Fixed Effects Model
A regression model with a dummy variable foreach individual in the sample, each observed Ti times.yi = Xi + diαi + εi, for each individual
1 1
2 2
N
=
=
N
1
2
N
y X d 0 0 0y X 0 d 0 0 β ε
αy X 0 0 0 d
β [X,D] εα Zδ ε
N may be thousands. I.e., the regression has thousands of variables (coefficients).
N columns
Part 4: Partial Regression and Correlation4-18/22
Estimating the Fixed Effects Model The FEM is a linear regression model but
with many independent variables
1
1Using the Frisch-Waugh theorem
=[ ]
D D
b XX XD Xya DX DD Dy
b XM X XM y
Part 4: Partial Regression and Correlation4-19/22
Fixed Effects Estimator (cont.)
i i i
i i i
i i i
1i
1 1 1T T T
1 1 1T T T
1 1 1T T T
(The dummy variables are orthogonal)
( ) (1/T)1- - ... -- 1- ... -... ... ... ...- - ... 1-
i i
1D
2D
D
ND
iD T i i i T i
M 0 00 M 0M
0 0 MM I d dd d =I dd
=
Part 4: Partial Regression and Correlation4-20/22
‘Within’ Transformations
i
i
Ti i i i t=1 it,k i.,k it,l i.,lk,l
Ti i i i t=1 it,k i.,k it i.k
, (x -x )(x -x )
, (x -x )(y -y )
N i iD i=1 D D
N i iD i=1 D D
XM X= XM X XM X
XM y= XM y XM y
Least Squares Dummy Variable Estimator
b is obtained by ‘within’ groups least squares (group mean deviations)