Hello all!
I would like to ask a question about building design matrixes. I know that there has been considerable effort in the mark-book (I have the new) to explain that in great detail, but I am not sure that I have understood everything to transfer the information to the following problem. (Apologizes in advance if this is an ‘avoidable’ question).
I have 3 age classes, the first (calves) includes the first interval only and is time dependent, the second class includes the second and third interval (juveniles), and from the 4th interval on I have the third age class (adults). The second and the third class are constant. This is so far my best model.
Now I want to test/analyze if I have a (linear) trend in that third (adult) class. Therefore I have to build this model via the design matrix. I have changed the PIM and design matrix as shown below. (I have more occasions/years than the shown 8 but it is just an example).
I would like to ask if the design matrix is correct.
The problem I have is that in the results I do get a trend but it is polynomial (-x2), which I do not understand – so my first idea was, that my design matrix might not be correct.
Many thanks in advance
Christian
I hoe you can read it - it looked better in excel
1 8 8 9 10 11 12
2 8 8 10 11 12
3 8 8 11 12
4 8 8 12
5 8 8
6 8
7
Itcpt Age Age Time1 T2 T3 T 4 T5 T6 T*A T*A T*A T*A T*A
1 1 0 1
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
1 1 0
1 0 1
1 0 0 1
1 0 0 2
1 0 0 3
1 0 0 4