Hanne, Are you having problems setting up your design matrix or getting TPSRegR to read the NTS file containing the design matrix?
For the first part I have written a very tiny SAS program to convert my varuables into the appropriate design matrix, which I can send to you. If it is the second problem, and you are using windows, I have experienced the problem and have a solution. After making your design matrix in excel or whatever program you use, save it first in a tab delimited format without the NTS header info i.e no: " File info 1 20 3L 0 CS S G Open the tab deimilited file in notepad (not wordpad) add your header info. Make sure that for your variables (CS S G in the above example) there is only a single space, and not a tab between each. Now save the file as such. File-> Save as Save as Type: All files Encoding: ANSI File name: whatever.nts This generally works for me. Cheers, Ian morphmet wrote: > I would like to post the following question on the mailing list (I'm new > on the list and apologize if the same question has come up several times > before!): > > > I have been advised to carry out a MANCOVA with CS as the covariate - > this should be the most straightforward way to compare shape in several > samples, while taking allometric growth effects into account. I have > tried to set up such an analysis in tpsRegr according to the example in > the help file. So far I have been unable to make a useable .nts file > with contrasts. > > If any one of you could tell me how to set up the .nts file for a > MANCOVA intpsRegr, would you please let me know on > [EMAIL PROTECTED] > > Thanks a lot in advance! > > Hanne Joergensen > > > ________________________________________________________________________ > Yahoo! for Good > Click here to donate to the Hurricane Katrina relief effort. > -- > Replies will be sent to the list. > For more information visit http://www.morphometrics.org > > -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
