Professor Miron Kaufman, Office SI-116, Tel.6872436,
Email:m.kaufman@popmail.csuohio.edu
Lectures: TH, 6:00PM-9:30PM, MC-317
Office Hour: TH 3:00PM-4:00PM, SI-116
Required Material: D.N.Gujarati, Basic Econometrics
Recommended Material: W.C.Beck Student Manual
The Quantitative Research Methods I is the first of a two-course sequence designed to provide Ph. D. students in Urban Studies with tools and skills necessary in quantitative research. Both courses focus on linear regression techniques. A good understanding will enable students to apply these techniques, as well as acquire on their own additional multivariate statistical techniques rooted in linear methodology, such as discriminant analysis and factor analysis.
In the first course in the sequence UST700 we study single-equation regression models with two and three variables, including estimation and inference. Part of this course is a project that spans the quantitative sequence. In UST700, a research problem is formulated and data are collected. Each student will identify a research problem, gather the data (more than two variables) on a diskette and write an essay of about 1000 words on this research problem. In UST701, students undertake the data analysis, stressing diagnosis and mitigation of problems related to data deviations from the basic linear regression assumptions.
Possible Data Sources:
- NODIS
- Gov’t Information Sharing Project, e.g., Earnings by Occupation & Education & Other data bases
- Federal Information Exchange
- Fedstats statistics from 70 agencies
- Census Bureau
- 1990 Census tables & Ferret, for recent population and income data
- National Criminal Justice Reference Service
- Criminal justice links
- Public Elementary-Secondary School Systems, Financial Statistics: Individual Unit Data
- Who Can Afford to Buy a House in 1993?
- Thomas, Library of Congress information service
- Bureau of Economic Analysis
- Other data sites
Part of UST700 is a computer lab where we will learn to work with the software MathCad which is useful for simulating and visualizing data sets. You should save your work on the diskette. At the end of the project each student will give me the diskette with all the programs and a printout of the results.
- TH 10/9/97 LAB#1: Tutorial and Two-Variable Regression
- TH 10/30/97 LAB#2: Monte Carlo Simulation of the Two-Variable Regression
- TH 11/20/97 LAB#3: Mean Prediction, Confidence Interval
- TH 11/20/97 LAB#4: Cobb-Douglas Production Function
TENTATIVE SCHEDULE
- Week #1 TH, 9/25: class organization, syllabus; Nature of Regression Analysis
Gujarati Intro., Ch.1.
- Week #2 TH, 10/2: Review of statistics concepts Gujarati Appendix A; Basics of two
variable regression model Gujarati Ch2; Hw.1 due.
- Week #3 TH, 10/9, Estimation in the two-variable regression model, Gujarati Ch.3;
Hw.2 due; Computer Lab. #1.
- Week #4 TH, 10/16, Normality Assumption, Gujarati Ch.4; Hw.3 due.
- Week #5 TH, 10/23, midterm exam
- Week #6 TH, 10/30, Inference in the two-variable regression model Gujarati Ch.5;
Project proposal due; Computer Lab. #2; F, 10/31, last day to drop class.
- Week #7 TH, 11/6, Extensions of the two-variable regression model Gujarati Ch.6;
Hw.4 due.
- Week #8 TH, 11/13, Estimation in the multiple regression model Gujarati Ch.7;
Hw.5 due.
- Week #9 TH, 11/20, Inference in the multiple regression model Gujarati Ch.8;
Computer Lab. #3; Hw.6 due.
- Week #10 TH, 11/27, NO CLASS, THANKSGIVING
- Week #11 TH, 12/4, Review; Project due.
- Exam week TH, 12/11 final exam
The final grade is a weighted average of:
- Computer Lab. 5%
- Homework 10%
- Project 20%
- Midterm Exam 30%
- Final Exam, Comprehensive 35%