Help... Time Series Problem...

[Dan Ryan]

[ Article crossposted from sci.stat.consult ]
[ Author was Dan Ryan ]
[ Posted on 3 Aug 1993 16:34:15 GMT ]

Hi... I have a group of 18 stationary timeseries, which in theory (based
upon the sampling scheme) should have the same covariance structure...
I would like to get an estimate of this covariance structure
using SAS if possible... I do most of my data analysis with SAS and
if I could use SAS, it would save alot of headaches.

Is there a way of tricking SAS into providing the MLE of the pooled
covariance structure using PROC ARIMA or any other procedure???

If not is there any other software out there which I could use???

If not how about some references on obtaining pooled estimates of
the covariance structure from multiple time series???

Any input anyone can provide will be greatly appreciated...

Dan Ryan
MTH...@VM.UOGUELPH.CA
DR...@SGI1.MATHSTAT.UOGUELPH.CA