Hi Marc
Try
VPREDICT [PRINT=desc, pred, se; PRED=predd; SE=see] CLASS=landuse, yr, yr2; \
LEVELS=*, !v(1,2,3,4,5,6,7,8,9,10,11,12,13,14), !v(1,4,9,16,25,36,49,64,81,100,121,144,169,196); PARALLEL=*,yr2, yr2
as this puts yr and yr2 in parallel. Previously you had landuse in parallel with yr, which was causing the problem.
Hope this helps,
Sue Welham
-----Original Message-----
From: GENSTAT-Request on behalf of Kery Marc
Sent: Fri 10/12/2004 11:08
To: [log in to unmask]
Cc:
Subject: VPREDICT (parameter PARALLEL)
Dear GenStatters,
please can you tell me what I am missing ? I am fitting a mixed model to hare counts over 14 years that were obtained at sites belonging to two landuse types. The model contains a linear and a quadratic year term. To get predictions for the 14 years and 2 landuse types, I did this:
VPREDICT [PRINT=desc, pred, se; PRED=predd; SE=see] CLASS=landuse, yr, yr2; \
LEVELS=*, !v(1,2,3,4,5,6,7,8,9,10,11,12,13,14), !v(1,4,9,16,25,36,49,64,81,100,121,144,169,196); PARALLEL=yr, yr2
On which GenStat gives the following fault:
**** G5F0061 **** Fault (Code VD 25). Statement 1 on Line 7280
Command: VPREDICT [PRINT=desc, pred, se; PRED=predd; SE=see] CLASS=landuse, yr, yr2; LEVELS=*, !v(1
Variables in PARALLEL must have same number of values
The two variables clearly both have 14 values, so this message is cryptic to me. Thanks a lot for your comments.
Kind regards --- Marc
__________________________________________________________
Marc Kery
Swiss Ornithological Institute
CH - 6204 Sempach
Switzerland
phone: 0041 41 462 97 93
email (work): [log in to unmask]
email (private): [log in to unmask]
http://www.unizh.ch/uwinst/homepages/kery.html
__________________________________________________________
|