The ARIMA Procedure

Forecasting Stage

To produce the forecast, use a FORECAST statement after the ESTIMATE statement for the model you decide is best. If the last model fit is not the best, then repeat the ESTIMATE statement for the best model before you use the FORECAST statement.

Suppose that the SALES series is monthly, that you want to forecast one year ahead from the most recently available SALES figure, and that the dates for the observations are given by a variable DATE in the input data set TEST. You use the following FORECAST statement:

   forecast lead=12 interval=month id=date out=results;

The LEAD= option specifies how many periods ahead to forecast (12 months, in this case). The ID= option specifies the ID variable, which is typically a SAS date, time, or datetime variable, used to date the observations of the SALES time series. The INTERVAL= option indicates that data are monthly and enables PROC ARIMA to extrapolate DATE values for forecast periods. The OUT= option writes the forecasts to the output data set RESULTS. See the section OUT= Data Set for information about the contents of the output data set.

By default, the FORECAST statement also prints and plots the forecast values, as shown in Figure 7.18 and Figure 7.19. The forecast table shows for each forecast period the observation number, forecast value, standard error estimate for the forecast value, and lower and upper limits for a 95% confidence interval for the forecast.

Figure 7.18: Forecasts for ARIMA(1,1,1) Model for SALES

The ARIMA Procedure

Forecasts for variable sales
Obs Forecast Std Error 95% Confidence Limits
101 171.0320 0.9508 169.1684 172.8955
102 174.7534 2.4168 170.0165 179.4903
103 177.7608 3.9879 169.9445 185.5770
104 180.2343 5.5658 169.3256 191.1430
105 182.3088 7.1033 168.3866 196.2310
106 184.0850 8.5789 167.2707 200.8993
107 185.6382 9.9841 166.0698 205.2066
108 187.0247 11.3173 164.8433 209.2061
109 188.2866 12.5807 163.6289 212.9443
110 189.4553 13.7784 162.4501 216.4605
111 190.5544 14.9153 161.3209 219.7879
112 191.6014 15.9964 160.2491 222.9538

Figure 7.19: Forecasts for the ARMA(1,1,1) Model

Forecasts for the ARMA(1,1,1) Model

Normally, you want the forecast values stored in an output data set, and you are not interested in seeing this printed list of the forecast. You can use the NOPRINT option in the FORECAST statement to suppress this output.