Class Notes for EC355
Ordinary Least Squares and Method of Moments
Derivation under the Gauss Markov assumptions
Jean Eid
Assume that the following hold
MLR1 Linearity in the parameter eg. suppose we have
y
=
β
0
+
β
1
x
1
+
β
2
x
2
+
β
3
x
3
+
···
+
β
k
x
k
+
u
MLR2 Independence of the error term u and random sampling
Independence of the error term means that the characteristics of individuals that we do not
observe are related to each others. A situation where this could be violated is as follow: Suppose
we survey children in Canadian households. Let’s say that we are interested in their educational
attainment, and we collect data on grade levels completed, age, and sex. You can imagine
a scenario where the more educated the child’s parents are, the higher his or her educational
attainment. Suppose we have a lot of brothers and sisters in our data. Since parents’ education
which is unobservable (we do not have data on) in our example have an effect on all their
children we end up having a situation were some of the unobserved characteristics (i.e. the us)
are correlated.
We need random sampling to represent the population, otherwise we cannot draw conclusion
for the whole population.
MLR3 No perfect Collinearity This means that we cannot obtain the values of one independent vari
able by forming linear combinations of some other ones. as an example, suppose we have the
following model.
y
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.
 Fall '11
 JEANEID
 Econometrics, Least Squares, Regression Analysis, Summation, Yi, i=1

Click to edit the document details