Business Forecasting Using SAS: A Point-and-Click Approach
Duration: 3.0 days CEU: 1.8
This course teaches business analysts how to carry out forecasting projects using the point-and-click interface in the Time Series Forecasting System in SAS/ETS software. Familiarity with business forecasting problems is recommended, but participants do not need forecasting or programming experience to benefit from this course.
Learn how to
- identify trend and seasonality in time series data
- propose simple trend, advanced trend, and/or seasonal models and fit these models to data
- identify input variables to include in a time series regression model
- identify the relationship between the target variable and input variables
- propose and fit dynamic time series regression models to data.
Who should attend
Forecasters and business analystsPrerequisites
Before attending this course you should be able to- create and access SAS data sets using the SAS windowing environment or a product such as SAS Enterprise Guide
- use software to explore and analyze data, including the ability to create graphical displays of data
- evaluate quantitative forecasts related to a business problem based on your advanced domain knowledge in an area such as finance, manufacturing, or retail.
A typical student will have college course work in a quantitative area, such as a basic business statistics course. Furthermore, a student should have experience using Microsoft Excel or a similar product for data summarization and presentation.
Course Contents
Business Forecasting- introduction to business forecasting
- introduction to time series forecasting
- measuring forecast accuracy
- descriptive and exploratory analysis of time series data
- modeling trend
- modeling seasonality
- using indicator variables to model events
- exponential smoothing models with trend components
- exponential smoothing models with trend and seasonal components
- introduction to Box-Jenkins forecasting
- autoregressive models
- moving average models
- mixed autoregressive moving average models
- identifying an appropriate autoregressive moving average model
- estimation and forecasting methods
- using differencing to model trend and seasonality
- trend models
- seasonal models
- forecasting the airline data
- forecasting the consumer electronics data
- ordinary regression models
- event models
- time series regression models
- event analysis using the seat belt law data
- event analysis using the airline data
- evaluating advertising effectiveness
- processing time-stamped data
- augmenting data with event variables