Date:         Thu, 21 Nov 2002 07:21:39 -0800
Reply-To:     jws@aggienetwork.com
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Warren Schlechte <warren.schlechte@ROCKETMAIL.COM>
Subject:      Need Nonparametric Paired Test for Serially Correlated Data
Content-Type: text/plain; charset=us-ascii

I wish to test whether one time series is significantly different (consistently
higher or lower by testing the medians) than is another.

My first thought is to use a Wilcoxon Sign Rank Test, as these data are not
normally distributed, but such a test requires independent samples.

Suggestions would be appreciated, especially those that are available in SAS.
Formulae for construction of tests not available in SAS would be welcome as
well.  References would be useful, but because of my location, are less useful
than the above two options.

Thanks for your thoughts and suggestions.


Warren Schlechte
HOH Fisheries Science Center
HC7 Box 62
Ingram, Texas 78025
Phone:   830.866.3356
Fax:       830.866.3549



__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus – Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com