Time Series Smoothing Models

[#1]

Use smoothing based time series models to describe patterns and forecast future time periods. 

Smoothing Models

1. From an open JMP^® data table, select Analyze > Specialized Modeling > Time Series.
2. Select a continuous variable from Select Columns, and click Y, Time Series (continuous variables have blue triangles). 
3. Select a time and click X, Time ID (optional). Click OK.  
   - A time series graph of the data along with a graph
     and table of the Auto and Partial Correlation values
     is displayed.
4. To fit a model to the data, click on the top red triangle, select Smoothing Model and choose the method. Here we choose Seasonal Exponential Smoothing.
5. Choose Observations per Period (e.g., 12) and click Estimate.
   - JMP displays the Model Summary, Parameter Estimates and a Forecast plot that shows the fit and forecasts from the model.
6. To estimate an alternative smoothing model, click on the red triangle, and choose another method (Winters Method was selected for a second model).
7. Click Estimate.
   - JMP provides a Model Comparison report, (below) that compares the two methods. Click and drag the slider bar at the bottom of the report to see all of the statistics.

In this example, the Seasonal Exponential Smoothing model fits the data better than Winters Method (i.e., AIC, SBC, MAPE, and MAE are smaller).

Workers.jmp (Help > Sample Data Folder > Time Series)

• The default number of forecast periods is 25. To change, enter a different value in the Time Series launch dialog window.
• To save the forecast, select Save Columns or Save Prediction Formula under the red triangle for that model.  A new table with the actual and predicted values will be generated.

Visit Predictive and Specialized Models > Time Series Analysis in JMP Help to learn more.