This preview shows page 1. Sign up to view the full content.
Econ 483. Computer Exercise 2DGP
Data are on the Catalyst Discussion Board (check myUW).
Data Generating Process and Least Squares Estimation. In observed studies, we never know what the
true model is. Therefore, it is di¢ cult to evaluate the estimation method such as least squares. In this
we have the data set, we pretend as if we do not know the parameters of the model and estimate them.
1. Download dgpols.dta and open it by using
use
command.
(1) How many variables are in the data set?
(2) What is the sample size?
2. The variable
x
is a collection of realized numbers from uniform
[0
;
2]
. The variable
u
is a collection of
realized normally distributed random numbers with mean zero and standard deviation 6.
(1) Calculate
Cov
(
X; U
)
.
(2) What is the sample covariance of
x
and
u
? [useful command: gen, sum]
(3) Is the sample covariance close to
Cov
(
X; U
)
? Can you say that
x
and
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '08
 Staff
 Macroeconomics

Click to edit the document details