serial correlation

Article View: bit.listserv.stat-l

Article #3604

Back to bit.listserv.stat-l Previous Next

#3604

From: PMTB00A
Date: Fri, 03 Jan 1992 16:44

10 lines
1101 bytes

The whole problem that you are having is that you have ignored the
underlying analytical structure. In particular if the disturbances u(t)
exhibit first order autocorrelation then we represent them by

                u(t) = p*u(t-1) + e(t)
where e(t) is serially independent. The variance of u(t) is an infinite
geometric series in var(e(t)). From this we obtain, the result that

                var(e(t)) = (1-p)*var(u(t))
Thus, your problem is solved with clear presentation.

Technical Details

Message-ID: <STAT-L%92010312450382@VM1.MCGILL.CA>

Path: arc1-ca.pugleaf.net!sparky!uunet!wupost!darwin.sura.net!paladin.american.edu!auvm!PANAM.BITNET!PMTB00A

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: f5efb,2c062ea27ea86a49,start
X-Google-Attributes: gidf5efb,public
X-Google-ArrivalTime: 1992-01-03 09:53:46 PST
Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
Path: arc1-ca.pugleaf.net!sparky!uunet!wupost!darwin.sura.net!paladin.american.edu!auvm!PANAM.BITNET!PMTB00A
Original_To:  BITNET%"stat-l@mcgill1"
Message-ID: <STAT-L%92010312450382@VM1.MCGILL.CA>
Newsgroups: bit.listserv.stat-l
Date:         Fri, 3 Jan 1992 11:44:00 CDT
Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
From:         PMTB00A@PANAM.BITNET
Subject:      serial correlation