Solutions to PS 3
*
Hongyan Zhao
Oct 15, 2009
Q1
(a) We reproduce the analysis described in the website and got the similar results (I
also uploaded the Matlab code to bspace).
*
If you find any mistake or typo, please email me [email protected]
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
(b) The correlation matrix among 10 questions is:
Q2
(a) The line didn’t fit the data well.
From the graph, we can see that the errors
variance are not constant. It violates the conditions that OLS estimator is BLUE. We
can use Whitetest to test whether there are heteroskedasticity.
(b) Using the fitted line, as age increases, earnings increases. But from the graph,
we don’t see the case. Instead, earnings increase with the age and hit a peak, then it
decreases. If the earnings of people under age 20 is zero (not the missing value), then
the residuals are just the negative of the fitted value. It is very big and impossible to
satisfy the mean zero conditions.
(c) We can use nonlinear regression function such as including square term into it.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '08
 Staff
 Linear Regression, Regression Analysis, Hongyan Zhao, ×V ar

Click to edit the document details