Hello
Apologies for any cross-posting.
I'm wishing to test the distribution or proportions along rows of
frequency data. The columns have a natural ordering. Its length
intervals. The rows also have a natural ordering as they are points in
time but its not so important that they are in chronological order.
I have fitted a log-linear model to the contingency table with main
effects only.
I would appreciate any thoughts and comments on how I've approached
this piece of work. And if anyone can suggest any alternatives?
Many thanks
Michelle
********************************************
Michelle Allen
Higher Scientific Officer
Biometrics Division
Department of Agriculture for Northern Ireland
Newforge Lane
Malone Road
Belfast BT9 5PX
E-mail: [log in to unmask]
Tel: (44) 028 90255215
Fax: (44) 028 90681216
********************************************
|