INSTRUCTOR: Ebenge Usip, Ph.D
OFFICE: 307 Debartolo Hall; Hrs. MWTTH 2-3pm; Ph. 742-1682; email: firstname.lastname@example.org
TEXT: Business Forecasting by Hanke, Wichern, & Reitsch, 2001, 7th ed.
Learning Economics & Business Statistics with SPSS/win (3rd ed.) by Usip and Handouts.
COURSE PREREQUISITES: Econ 2610 & 3780
GRADING WEIGHT& SCALE:
Exam # 1 (Date to be announced) 20% of 500 points
Exam # 2 (Date to be announced) 20% of 500 points
Exam # 3 (Date to be announced) 20% of 500 points
Homework 10% of 500 points
Final Exam: () 30% of 500 points
APPROXIMATE SCALE: Will be explained in class
Note: Last Day to WITHDRAW with a "W" is Oct. 27 (by noon)
Part 1: Fundamentals of Time Series Analysis and Forecasting
A. Introduction to Forecasting: chapter 1
B. A Review of Basic Statistical Concepts: chapter 2
C. Exploring Data Patterns [using Time Series Plot (TSP) & Correlogram]: chapter 3, pp. 53 - 69
D. Choosing a Forecasting Technique: chapter 3, pp. 69 - 79
E. Time Series and Their Components (selected topics): chapter 5
Part 2: Forecasting with Moving Averages & Smoothing Methods: Chapter 4
A. The Moving Average Models
B. The Simple or Single Exponential Smoothing (SES) Model
C. The Holt Exponential Smoothing (HES) Model
D. The Winters Exponential Smoothing (WES) Model
MIDTERM # 2
Part 4: Forecasting with Regression Methods
A. Forecasting with Simple Regression Models: chapter 6
B. Forecasting with Multiple Regression and Econometrics Models: chapters 7 & 8
MIDTERM # 3
Part 5: Forecasting with Box-Jenkins (ARIMA) Methodology & Managing Forecasts
A. Forecasting with Non-seasonal ARIMA models: chapter 9, pp 346 - 378
B. Forecasting with Seasonal ARIMA models: chapter 9, pp. 379 - 390
C. Judgment Elements in Forecasting: chapter 10
D. Managing the Forecasting Process: chapter 11
FINAL EXAM (Dec. 11, 1515 - 1715)
1. Class attendance is highly recommended. No make-up exams will be administered.
Late homework will not be graded.
2. It is the student's responsibility to be familiar with the assigned readings and all the
materials covered during lectures. Class participation is encouraged and rewarded.
3. Use of SPSS/win and Web tutorials (Learning Statistics with SPSS/win ) are the required part of the
OVERVIEW: Goals and
Forecasting the future is a fundamental aspect of decision making in any business or government. Since economic and business conditions vary over time, business and government leaders must find ways to keep abreast with the effects that such changes will have on their operations. For instance, a business executive is especially concerned with such key decision variables as the future sales, profits, stock prices; while a government official may worry about the future rates of inflation/unemployment, and the levels (time paths) of the GNP. Time-series Forecasting is a set of quantitative techniques that have proved useful in planning the future needs and controlling present operations.
The objective of this course is two-fold. First, to familiarize you with the many statistical techniques that are useful in the forecasting of economic and business time-series. Second, to expose you to a variety of applications in the form of problems and cases. This unified approach (minimal theory and more emphasis on applications) is intended to help you develop a pattern of thought that will persist after you enter the business world as a time-series analyst. No great skill in mathematical statistics is required beyond introductory statistics, Econ. 3780 or the equivalent. You must however be prepared to adapt to new statistical concepts and symbols pertinent to the discussion of time-series analysis and forecasting.
Many of the procedures that we will examine involve messy calculations. In today's computerized environment, the optimal focus in teaching a quantitative course such as this places less emphasis on hand computation and more on concepts. To this end, computer application is an integral part of this course. As much as possible, we will use the SPSS/win statistical program to implement the forecasting techniques that are presented in the text. But remember that SPSS/win is not a dedicated econometric/forecasting program such as Eviews, which the authors recommend For instance, SPSS/win does not automatically compute measures of forecast adequacy such as MAD, MSE, RMSE, MAPE, MPE, and Ljung-Box Q statistic. Thus, in some problems and cases limited manual computation will be required using the results from the SPSS/win outputs.
STUDY HINTS: You will enjoy this course if you are
willing to put in a minimum amount of effort. Unlike many
quantitative economics courses, most of the materials are pretty
straightforward and have immediate bearing to reality. People who
get too far behind probably won't be able to catch up. Be sure
to read the assignments before coming to class, come to class
regularly, and come by my office at the first sign of trouble.
Have a successful semester.
Top or Back to Course Overview, or Course Information, or Home Page