LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (November 2002, week 2)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Tue, 12 Nov 2002 16:33:10 +0530
Reply-To:   sahu prasanta <Sahu.Prasanta@UNILEVER.COM>
Sender:   "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:   sahu prasanta <Sahu.Prasanta@UNILEVER.COM>
Subject:   Multiple comparision test using LSD
Comments:   cc: "Rao, Girish" <>

Hi all, This is not a SAS question but a general programming query for statisticians among the group. We are writing an EXCEL macro for doing a Multiple comparision test. I have got a set of means and a LSD value from a proc glm analysis.I want to compare some more set of means using this LSD so that the output will be in the same way as proc glm gives with option "means a b/lsd" would give i.e, means with same letters are not significantly different. We know the logic of assigning letters to means, i.e, I start with max/min value and compare it with successive observations and keep giving the same letter untill the difference becomes more than LSD, and stop once the last mean is counted. We are struggling with the codes and not able to impelement the logic properly. I know this question is not exactly meant for the group like this one but, I don't know any other group where there are good programmers with statistical background. So any code in C or VB or SAS will be enough. Putting as a SAS question, the problem would look something like this to me. I have got the set of sorted means as variable1 in a data set. Using the LSD value of say "0.4", I want to compare the values in variable1 starting from _n_=1 and fill variable2 with letters a,b,c and so on as proc glm would do it using the above logic. Thank you for your help With regards Prasanta sahu

Back to: Top of message | Previous page | Main SAS-L page