Using Diagnostics to Identify ARIMA Models

/*--------------------------------------------------------------

                    SAS Sample Library

        Name: ariex05.sas
 Description: Example program from SAS/ETS User's Guide,
              The ARIMA Procedure
       Title: Using Diagnostics to Identify ARIMA Models
     Product: SAS/ETS Software
        Keys: time series analysis
        PROC: ARIMA
       Notes:

--------------------------------------------------------------*/


title1 'SERIES A: Chemical Process Concentration Readings';
data SeriesA;
   n=_n_;
   input x :3.1 @@;
datalines;
170 166 163 161 171 169 168 174 171 170
167 174 172 174 174 170 173 172 174 168
171 174 174 175 174 176 174 173 170 178
175 181 175 174 174 171 176 177 174 178
176 175 165 178 173 173 171 174 169 173
176 169 167 168 168 172 168 176 172 166
171 169 166 180 172 173 170 169 173 168
173 174 177 168 169 170 169 170 166 167
168 167 164 165 164 166 165 167 164 164
162 164 163 164 170 169 171 171 167 169
165 172 164 170 170 167 162 166 169 165
166 166 170 171 171 167 168 163 166 168
169 171 168 170 172 173 172 173 172 172
175 169 169 169 170 165 167 168 167 167
166 165 170 167 167 169 174 171 170 168
172 172 174 172 169 168 170 174 172 172
171 171 171 174 172 169 169 170 167 169
173 178 178 176 175 170 169 171 172 174
175 179 170 170 170 172 173 174 174 170
180 182 176 178 177 172 174
;

/*-- Order Identification Diagnostic with SCAN Method --*/
proc arima data=SeriesA;
   identify var=x scan;
run;

/*-- Order Identification Diagnostic with ESACF Method --*/
proc arima data=SeriesA;
   identify var=x esacf;
run;

/*-- Combination of SCAN and ESACF Methods --*/
proc arima data=SeriesA;
   identify var=x scan esacf;
run;

/*-- Augmented Dickey-Fuller Unit Root Tests --*/
proc arima data=SeriesA;
   identify var=x stationarity=(adf=(5,6,7,8));
run;

/*-- Minimum Information Criterion --*/
proc arima data=SeriesA;
   identify var=x(1) minic;
run;

/*-- Combination of MINIC, SCAN and ESACF Options --*/
proc arima data=SeriesA;
   identify var=x(1) minic scan esacf;
run;

title;