2662 ✦ Chapter 39: Getting Started with Time Series Forecasting Figure 39.44 Model Viewer: Parameter Estimates Table For the linear trend model, the parameters are the intercept and slope coefficients. The table shows the values of the fitted coefficients together with standard errors and t tests for the statistical significance of the estimates. The model residual variance is also shown. Statistics of Fit Table Select the sixth icon from the top in the vertical toolbar to the right of the table. This switches the Viewer to display a table of statistics of fit computed from the model prediction errors, as shown in Figure 39.45. The list of statistics displayed is controlled by selecting Statistics of Fit from the Options menu. Changing to a Different Model ✦ 2663 Figure 39.45 Model Viewer: Statistics of Fit Table Changing to a Different Model Select the first icon in the vertical toolbar to the right of the table to return the display to the predicted and actual values plots (Figure 39.39). Now return to the Develop Models window, but do not close the Model Viewer window. You can use the Next Viewer icon in the toolbar or your system’s window manager controls to switch windows. You can resize the windows to make them both visible. Select the Double Exponential Smoothing model so that this line of the model list is highlighted. The Model Viewer window is now updated to display a plot of the predicted values for the Double Exponential Smoothing model, as shown in Figure 39.46. The Model Viewer is automatically updated to display the currently selected model, unless you specify Unlink (the third icon in the window’s horizontal toolbar). 2664 ✦ Chapter 39: Getting Started with Time Series Forecasting Figure 39.46 Model Viewer Plot for Exponential Smoothing Model Forecasts and Confidence Limits Plots Select the seventh icon from the top in the vertical toolbar to the right of the graph. This switches the Viewer to display a plot of forecast values and confidence limits, together with actual values and one-step-ahead within-sample predictions, as shown in Figure 39.47. Data Table ✦ 2665 Figure 39.47 Model Viewer: Forecasts and Confidence Limits Data Table Select the last icon at the bottom of the vertical toolbar to the right of the graph. This switches the Viewer to display the forecast data set as a table, as shown in Figure 39.48. 2666 ✦ Chapter 39: Getting Started with Time Series Forecasting Figure 39.48 Model Viewer: Forecast Data Table To view the full data set, use the vertical and horizontal scroll bars on the data table or enlarge the window. Closing the Model Viewer Other features of the Model Viewer and Develop Models window are discussed later in this book. For now, close the Model Viewer window and return to the Time Series Forecasting window. To close the Model Viewer window, select Close from the window’s horizontal toolbar or from the File menu. Chapter 40 Creating Time ID Variables Contents Creating a Time ID Value from a Starting Date and Frequency . . . . . . . . . . . . 2667 Using Observation Numbers as the Time ID . . . . . . . . . . . . . . . . . . . . . . 2671 Creating a Time ID from Other Dating Variables . . . . . . . . . . . . . . . . . . 2674 The Forecasting System requires that the input data set contain a time ID variable. If the data you want to forecast are not in this form, you can use features of the Forecasting System to help you add time ID variables to your data set. This chapter shows examples of how to use these features. Creating a Time ID Value from a Starting Date and Frequency As a first example of adding a time ID variable, use the SAS data set created by the following statements. (Or use your own data set if you prefer.) data no_id; input y @@; datalines; 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 run; Submit these SAS statements to create the data set NO_ID. This data set contains the single variable Y. Assume that Y is a quarterly series and starts in the first quarter of 1991. In the Time Series Forecasting window, use the Browse button to the right of the Data set field to bring up the Data Set Selection window. Select the WORK library, and then select the NO_ID data set. You must create a time ID variable for the data set. Click the Create button to the right of the Time ID field. This opens a menu of choices for creating the Time ID variable, as shown in Figure 40.1. 2668 ✦ Chapter 40: Creating Time ID Variables Figure 40.1 Time ID Creation Popup Menu Select the first choice, Create from starting date and frequency . This opens the Time ID Creation from Starting Date window shown in Figure 40.2. Creating a Time ID Value from a Starting Date and Frequency ✦ 2669 Figure 40.2 Time ID Creation from Starting Date Window Enter the starting date, 1991:1, in the Starting Date field. Select the Interval list arrow and select QTR. The Interval value QTR means that the time interval between successive observations is a quarter of a year; that is, the data frequency is quarterly. Now select the OK button. The system prompts you for the name of the new data set. If you want to create a new copy of the input data set with the DATE variable added, enter a name for the new data set. If you want to replace the NO_ID data set with the new copy containing DATE, just select the OK button without changing the name. For this example, change the New name field to WITH_ID and select the OK button. The data set WITH_ID is created containing the series Y from NO_ID and the added ID variable DATE. The system returns to the Data Set Selection window, which now appears as shown in Figure 40.3. 2670 ✦ Chapter 40: Creating Time ID Variables Figure 40.3 Data Set Selection Window after Creating Time ID Select the Table button to see the new data set WITH_ID. This opens a VIEWTABLE window for the data set WITH_ID, as shown in Figure 40.4. Select File and Close to close the VIEWTABLE window. Using Observation Numbers as the Time ID ✦ 2671 Figure 40.4 Viewtable Display of Data Set with Time ID Added Using Observation Numbers as the Time ID Normally, the time ID variable contains date values. If you do not want to have dates associated with your forecasts, you can also use observation numbers as time ID variables. However, you still must have an ID variable. This can be illustrated by adding an observation index time ID variable to the data set NO_ID. In the Data Set Selection window, select the data set NO_ID again. Select the Create button to the right of the Time ID field. Select the fourth choice, Create from observation numbers . This opens the Time ID Variable Creation window shown in Figure 40.5. . display the forecast data set as a table, as shown in Figure 39. 48. 2666 ✦ Chapter 39: Getting Started with Time Series Forecasting Figure 39. 48 Model Viewer: Forecast Data Table To view the full data. Time ID Value from a Starting Date and Frequency ✦ 26 69 Figure 40.2 Time ID Creation from Starting Date Window Enter the starting date, 199 1:1, in the Starting Date field. Select the Interval list. the right of the table to return the display to the predicted and actual values plots (Figure 39. 39) . Now return to the Develop Models window, but do not close the Model Viewer window. You can