Problem Set 1: Simple Linear Regression
Due: Thursday, February 12, in class. You must show all your work to get full
1. Regression with Logs. Using data from 1988 for houses sold in Andover, Massachusetts, the model that relates housing
Problem Set 3: Multiple Linear Regression
Due: Tuesday, March 17th, in class. When not species, use 5% signicance level. You
must show all your work to get full credit.
1. Consider the multiple linear regression model with three independent varia
Problem Set 2: Multiple Linear Regression
Due: Thursday, March 5, in class. You must show all your work to get full credit.
1. A researcher wants to study the trade-o between time spent sleeping and time spent
working, so he plans to estimate the
2.3 Theory behind regression:
Assume we define our population as: three person household in American. We define x as weekly
income in units of 100$ and the values of y are weekly food expenditures. We are interested how y
varies with changes in x.
Choose 30 states randomly, google these states populations and GDP, by regression in SAS check if
there is a relation between states populations and GDP. (I believe the best source is Wikipedia)
(Actually we have GSP for states but I still used the word G
NORTHERN ILLINOIS UNIVERSITY
DEPARTMENT OF ECONOMICS
ECON 390A - 001
INSTRUCTOR: Shawn Osell
Office Hours: ZU 502 Monday and Wednesday: 2:00 - 3:00PM
(Note: I will usually be in my office throughout the week or DU 470).