| Date: | Fri, 6 Feb 2004 10:41:43 -0600 |
|---|
| Reply-To: | Michal Lijowski <michal@CVU.WUSTL.EDU> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Michal Lijowski <michal@CVU.WUSTL.EDU> |
|---|
| Organization: | Washington University in St. Louis |
|---|
| Subject: | Using PROC MODEL and FIT |
|---|
| Content-Type: | text/plain; charset=us-ascii; format=flowed |
|---|
How do I incorporate uncertainties into the FIT statement?
Below is a code taken from A Study of Kinetics: The Estimation
and Simulation of Systems of First-Order Differential Equation
by D. Erdman and M.M. Morelock and I added 5% errors
to the conc data. How do I include those errors in the FIT statement?
Michal
data fish;
input day conc concerr;
datalines;
0.0 0.0 0.0
1.0000 0.1352 0.00676
2.0000 0.2168 0.01084
3.0000 0.255 0.01275
4.0000 0.3258 0.01629
6.0000 0.3313 0.016565
;
run;
proc model data = fish;
parm ku 0.3 ke 0.3 conc0 0;
dert.conc = ku - ke * conc;
fit conc initial = (conc = conc0) / time = day dynamic;
run;
| 
