turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- SAS Procedures
- /
- nlmixed - random effect covariate calculation and ...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

04-02-2014 03:55 PM

Hi,

I have a similar problem as one of the earlier post(with 3 random variables), , repeated measure of performance over 4-5 trials. I have a group variable (ordinal - 3 groups).

I'd like to test the differences between the groups (1 vs 2, 1 vs 3, and 2 vs 3).

Here is my attempt to the code,

proc nlmixed data=long;

parms a1=1 b1=.5 c1=.5 u1=.1 u2=.1 u3=.1 s2e=?? g11=?? g21=?? ...

a2=0 a3=0 b2=0 b3=0 c2=0 c3=0;

a=a1+a2*(group=2)+a3*(group=3)+u1;

b=b1+b2*(group=2)+b3*(group=3)+u2;

c=c1+c2*(group=2)+c3*(group=3)+u3;

eqn=a/time**b+c;

model performance ~ normal(eqn,s2e);

random u1 u2 u3 ~ normal([0,0,0],[g11,g21,g11,g31,g32,g33]) subject=id;

run;

2 things I am having trouble with are:

1. how to code "CONTRAST" to test the proper groups. is it just

contrast "b1 vs b2" b1 - b2;

contrast "b1 vs b3" b1 - b3;

contrast "b2 vs b3" b2 - b3;

contrast "a1 vs a2" a1 - a2;

....

2. how do I calculate the values to s2e, g11, g21....?

thanks in advance for your help.

ming

Accepted Solutions

Solution

04-03-2014
08:31 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

04-03-2014 08:31 AM

1. It looks like these should work, but it may be worthwhile to consider using ESTIMATE statement due to the additional info it gives.

2. Coming up with initial values, especially for covariances, is hard. I would try the default values and see where that leads. Then I would go back and refit with different values, roughly based on the results, and see if I got comparable convergence to the same values. If you have a lot of time, you could try the grid searching option.

Another method would be to semi-linearize, and use PROC MIXED to get approximate starting values. Subtract your best common estimate of c from performance, take the logs of both sides, and fit a regression model that accomodates the random effects and the values from the V matrix.

Steve Denham

All Replies

Solution

04-03-2014
08:31 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

04-03-2014 08:31 AM

1. It looks like these should work, but it may be worthwhile to consider using ESTIMATE statement due to the additional info it gives.

2. Coming up with initial values, especially for covariances, is hard. I would try the default values and see where that leads. Then I would go back and refit with different values, roughly based on the results, and see if I got comparable convergence to the same values. If you have a lot of time, you could try the grid searching option.

Another method would be to semi-linearize, and use PROC MIXED to get approximate starting values. Subtract your best common estimate of c from performance, take the logs of both sides, and fit a regression model that accomodates the random effects and the values from the V matrix.

Steve Denham