BADM 675 Business Forecasting, Fall 2002
Dr. S. Silver Office: Bond Hall
273: HRS: M
Tel: 953-5163; Home 766-9943; E-mail: silvers@citadel.edu
COURSE DESCRIPTION
Forecasting is the science, and art, of predicting the future. For business people the ability to make accurate forecasts may be the difference between failure and survival in an ever increasingly competitive environment.
The course will include coverage of some of the most commonly used
techniques by business. These include smoothing techniques, regression
analysis, times series decomposition, time series analysis (ARIMA models), and
consensus forecasts. The first half of
the course will concentrate on a review of statistics and estimation
techniques, particularly multivariate regression analysis. The exam covering the material from this part
of the course will be worth 40% of the final grade.
The second half will focus on forecasting theory and practice. We will study various methods to translate model estimates into forecasts and various univariate forecasting techniques including exponential smoothing, smoothing with moving averages, classical time series decomposition, and time series analysis. The last of these, TSA, will be for discussion only. Even though it is commonly used by practitioners, in practice it is very easily misused and misinterpreted. The second half of the course will be grades based on a forecasting project, to be designed, prepared, and presented by each student. The brief presentation will be done the last Saturday class and will be worth 40% of the final grade.
Students will also use the various techniques studied to produce their own forecasts of a time series chosen be the instructor. Winner of the contest will compete in a contest for "Forecaster of the Semester", rules of which will be explained later.
TEXTS AND OTHER READINGS
Required text is DeLurgio, Stephen A., Forecasting Principles and Applications, (1998, Irwin McGraw-Hill).
GRADING
In addition to a midterm and a non-comprehensive final, there will be several homework/mini-projects, many of which will require the use of the computer, and a grade for the "Forecaster of the Semester" contest. Grading will be approximately as follows:
|
Exam 1 |
35 % |
|
Exam 2 |
35 % |
|
Homework/projects |
20 % |
|
Forecaster of the Semester |
10 % |
Organization of lectures -- Lecture topics and chapters.
|
Week |
Topic |
Chapters |
|
1 |
Introduction to forecasting-planning and forecasting |
Chapter 1 |
|
2 |
Statistical foundations of forecasting Handout |
Chapter 2 |
|
3 |
Simple linear
regression model and correlation |
Chapter 3 |
|
4 |
The multiple
regression model |
Chapter 10 |
|
5 |
Econometric methods |
Chapter 11 |
|
6 |
More on multiple
regression models |
Handout |
|
7 |
Exam 1 Chapters 1-3, 10, 11 |
|
|
10 |
Smoothing
techniques and models |
Chapter 4 |
|
11 |
Time series
decomposition |
Chapter 5 |
|
12 |
Trend-seasonal and Holt-Winter Smoothing |
Chapter 6 |
|
8 |
Time series analysis; ARIMA models; autocorrelation and ACFs |
Chapter 7 |
|
9 |
Time series analysis; ARIMA models and applications |
Chapters 7 - 9 |
|
13 |
Cyclical forecasts |
Chapter 14 |
|
14 |
Technological and qualitative forecasts |
Chapter 15 |
|
15 |
Exam 2 Chapters 4-9, 14, 15 |
|