Multi-period Autocorrelation P-Values

[harryg]

Given a time series dataset, is there a SAS procedure that can easily generate the autocorrelations / lagged correlations for multiple periods (12) and the associated p-values in tabular format?

[udo_sas]

Hello -

If you have access to SAS/ETS software then you might want to have a look at the TIMESERIES procedure, which provides access to both autocorrelation and crosscorrelation features.

See:

http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_timeseries_sec...

http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_timeseries_sec...

Example (autocorrelation):

data monthly;                                                                                                                          

set monthlydata_sas7;                                                                                                                  

sasdate=input(put(month,z6.),yymmn6.);                                                                                                 

drop month;                                                                                                                            

format sasdate date9.;                                                                                                                 

run;                                                                                                                                   

proc timeseries data=monthly out=_null_ outcorr=outcorr;                                                                               

id sasdate interval=month;                                                                                                             

var sprd;                                                                                                                              

corr / nlag=12;                                                                                                                        

run;

proc print noobs data=outcorr;

run;

Thanks,

Udo