Introduction to Business Forecasting: Using ForecastX™ |
|
This web-based course is designed for working professionals and university students who want to enhance their understanding of business forecasting from an applied perspective. The course explains how forecasts are actually developed and utilized, emphasizing modern statistical methods and quantitative forecasting techniques. Specific applications to business include forecasting sales, production, inventory, macroeconomic factors such as interest rates and exchange rates, and other aspects of both short- and long-term business planning.
Course Specifications: |
 |
|
Length: |
|
10 weeks |
|
|
|
Weekly time commitment: |
|
Approximately 5 hours |
|
|
|
Prerequisites: |
|
None, however familiarity with Microsoft Excel, basic algebra and basic statistics is preferred. |
|
|
|
Tuition: |
|
US $1,495 |
|
|
|
Additional expenses: |
|
Textbook including ForecastX™ software.
U.S. Edition is approximately $125.
Foreign Edition prices vary by location. |
|
|
Course Materials - Text and Software
The course is based upon the text, Business Forecasting, by J. Holton Wilson and Barry Keating in cooperation with John Galt Solutions, Inc. Published by McGraw-Hill Higher Education, the text is in its 5th Edition. The course utilizes Microsoft Excel-based ForecastX™ Software for Windows, which accompanies the text. ForecastX™ is one of the most comprehensive, Excel-based forecasting tools available on the market
The text can be purchased directly from McGraw-Hill Higher Education at http://www.mhhe.com/. The ISBN is 0-07-320398-X.
System Requirements
The minimum system requirements necessary to complete the course are as follows:
- A 486X-based computer
- Microsoft Windows 95 or later or Windows NT 4.0 or later.
- A minimum of 8MB of memory with 32MB recommended
- Microsoft Office 97 or later is required to run the ForecastX™ Wizard
- Access to the Internet and an email account with the ability to send and receive large Microsoft Excel and Word files
Course Format
The course runs 10 weeks and consists of 9 weekly units. A final 10th week is provided to complete an independent forecast implementation project.
Each week the student will read a chapter in the text and the associated lecture notes and complete assigned exercises using the ForecastX™ software. A weekly quiz will enable the student to test his or her knowledge of the chapter material. The independent project will allow the student to apply the forecasting techniques they have learned to a real world situation of personal or professional interest.
Course Content
Unit 1: Introduction to Business Forecasting
The first unit introduces the student to the role of quantitative forecasting in business, public and non-profit sectors. While the focus is on quantitative techniques, a brief discussion of subjective forecasting methods and new product forecasting is provided. Two naïve forecasting models are introduced and methods for evaluating model accuracy are discussed in detail. Two ongoing case studies are introduced, which allow the student to apply the techniques covered in the chapter to real world situations: forecasting domestic car sales and forecasting sales for The Gap.
Unit 2: The Forecast Process, Data Considerations, and Model Selection
Unit two provides an introduction to the nine-step forecast process. Data patterns are discussed including trend, seasonality and cycle and their implications for model selection. A review of the statistical concepts used in forecasting is provided. Case studies on domestic car sales and The Gap sales continue.
Unit 3: Moving Averages and Exponential Smoothing
This unit introduces moving averages and exponential smoothing techniques. Specifically, simple exponential smoothing, Holt's exponential smoothing, Winters' exponential smoothing, and adaptive response rate single exponential smoothing techniques are explained in detail. The event modeling capabilities of FORECAST™ are briefly demonstrated. Case studies on domestic car sales and The Gap sales continue.
Unit 4: Introduction to Forecasting with Regression Methods
This unit introduces the Bivariate Regression Model. A process for generating forecasts using regression methods is discussed including simple linear trend models and causal regression models. The statistical evaluation of regression model results is discussed. The application of the standard error of the estimate in generating approximate confidence intervals is detailed. Problems associated with regression analysis are introduced including serial correlation and heteroscedasticity. Cross-sectional forecasting techniques are introduced. Case studies on domestic car sales and The Gap sales continue.
Unit 5: Forecasting with Multiple Regression
Unit 5 extends the introduction to forecasting with regression methods developed in Chapter 4 to include the multiple regression model. Selecting independent variables is discussed, as is the statistical evaluation of multiple regression models. The problem of serial correlation is again addressed and alternative variable selection criteria are introduced. The use of dummy variables is introduced as an effective method to account for seasonality or other qualitative attributes of the data. A brief discussion of the use of nonlinear terms in the regression model is provided. Case studies on domestic car sales and The Gap sales continue.
Unit 6: Time Series Decomposition
Unit 6 introduces the classic time-series decomposition model and discusses its popularity among business managers. Methods for deseasonalizing the data and finding seasonal indexes are demonstrated, as are techniques for finding the long-term trend and measuring the cyclical component in the data. Finally, the time-series components are recombined to generate the time-series decomposition forecast. Case studies on domestic car sales and The Gap sales continue.
Unit 7: ARIMA (Box-Jenkins) Type Forecasting Models
The philosophy of Box-Jenkins is outlined in Unit 7 as an introduction to the methodology of ARIMA (Box-Jenkins) forecasting models. Moving average, autoregressive and mixed autoregressive - moving average models are discussed. Stationarity and methods for removing non-stationarity in time-series data are discussed in detail. Finally, the Box-Jenkins identification process is defined and demonstrated. Case studies on domestic car sales and The Gap sales continue.
Unit 8: Combining Forecast Results
Unit 8 introduces the techniques and benefits of combining forecast results. The problem of forecast bias is discussed. Three techniques for selecting the appropriate weights for combined forecasts are detailed. Case studies on domestic car sales and The Gap sales continue.
Unit 9: Forecast Implementation
The final chapter of the text details the forecast implementation process from an applied perspective. Practical keys to obtaining better forecasts are provided. Guidelines for selecting the appropriate forecasting technique are provided. New product forecasting techniques are discussed. The text ends with an introduction to artificial intelligence and applications to forecasting.
Independent Forecast Implementation Project
At the beginning of the term, students will select a topic of personal or professional interest and on a weekly basis, apply the new forecasting techniques learned to their topic. At the end of the term, students will generate a final forecast based upon the nine-step forecasting process. The 10th week of the course will be used for students to complete the project. Students will submit a brief report of approximately 10 pages detailing their forecast process and results.
Comments or Questions?
Register now.
|
