Introduction to Time Series Analysis Using IBM SPSS Modeler (v18.1.1)

Audience

Roles: Business Analyst, Data Scientist

Specifically, this is an introductory course for:

• Anyone who is interested in getting up to speed quickly and efficiently using the IBM SPSS Modeler forecasting capabilities

Prerequisites

• Familiarity with the IBM SPSS Modeler environment (creating, editing, opening, and saving streams).
• General knowledge of regression analysis is recommended but not required

Course topics

1: Introduction to time series analysis

• Explain what a time series analysis is
• Describe how time series models work
• Demonstrate the main principles behind a time series forecasting model

2: Automatic forecasting with the Expert Modeler

• Examine fit and error
• Examine unexplained variation
• Examine how the Expert Modeler chooses the best fitting time series model

3: Measuring model performance

• Discuss various ways to evaluate model performance
• Evaluate model performance of an ARIMA model
• Test a model using a holdout sample

4: Time series regression

• Use regression to fit a model with trend, seasonality and predictors
• Handling predictors in time series analysis
• Detect and adjust the model for autocorrelation
• Use a regression model to forecast future values

5: Exponential smoothing models

• Types of exponential smoothing models
• Create a custom exponential smoothing model
• Forecast future values with exponential smoothing
• Validate an exponential smoothing model with future data

6: ARIMA modeling

• Explain what ARIMA is
• Learn how to identify ARIMA model types
• Use sequence charts and autocorrelation plots to manually identify an ARIMA model that fits the data
• Check your results with the Expert Modeler