Chapter Contents
Chapter Contents
Previous
Previous
Next
Next
The FORECAST Procedure

Functional Summary

The statements and options controlling the FORECAST procedure are summarized in the following table:

Description Statement Option
Statements   
specify BY-group processingBY 
identify observationsID 
specify the variables to forecastVAR 
   
Input Data Set Options   
specify the input SAS data setPROC FORECASTDATA=
specify frequency of the input time seriesPROC FORECASTINTERVAL=
specify increment between observationsPROC FORECASTINTPER=
specify seasonalityPROC FORECASTSEASONS=
specify number of periods in a seasonPROC FORECASTSINTPER=
treat zeros at beginning of series as missingPROC FORECASTZEROMISS
   
Output Data Set Options   
specify the number of periods ahead to forecastPROC FORECASTLEAD=
name output data set containing the forecastsPROC FORECASTOUT=
write actual values to the OUT= data setPROC FORECASTOUTACTUAL
write confidence limits to the OUT= data setPROC FORECASTOUTLIMIT
write residuals to the OUT= data setPROC FORECASTOUTRESID
write standard errors of the forecasts to the OUT= data setPROC FORECASTOUTSTD
write one-step-ahead predicted values to the OUT= data setPROC FORECASTOUT1STEP
write predicted, actual, and confidence limit values to the OUT= data setPROC FORECASTOUTFULL
write all available results to the OUT= data setPROC FORECASTOUTALL
specify significance level for confidence limitsPROC FORECASTALPHA=
control the alignment of SAS Date valuesPROC FORECASTALIGN=
   
Parameters and Statistics Output Data Set Options   
write parameter estimates and goodness-of-fit statistics to an output data setPROC FORECASTOUTEST=
write additional statistics to OUTEST= data setPROC FORECASTOUTESTALL
write Theil statistics to OUTEST= data setPROC FORECASTOUTESTTHEIL
write forecast accuracy statistics to OUTEST= data setPROC FORECASTOUTFITSTATS
   
Forecasting Method Options   
specify the forecasting methodPROC FORECASTMETHOD=
specify degree of the time trend modelPROC FORECASTTREND=
specify smoothing weightsPROC FORECASTWEIGHT=
specify order of the autoregressive modelPROC FORECASTAR=
specify significance level for adding AR lagsPROC FORECASTSLENTRY=
specify significance level for keeping AR lagsPROC FORECASTSLSTAY=
start forecasting before the end of dataPROC FORECASTSTART=
specify criterion for judging singularityPROC FORECASTSINGULAR=
   
Initializing Smoothed Values   
specify number of beginning values to use in calculating starting valuesPROC FORECASTNSTART=
specify number of beginning values to use in calculating initial seasonal parametersPROC FORECASTNSSTART=
specify starting values for constant termPROC FORECASTASTART=
specify starting values for linear trendPROC FORECASTBSTART=
specify starting values for the quadratic trendPROC FORECASTCSTART=
   

Chapter Contents
Chapter Contents
Previous
Previous
Next
Next
Top
Top

Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.