Hi everyone,
I'm running proc NLIN to output parameter estimates, and would like to add the standard error of the parameter estimates to the output dataset. Proc REG has the outseb option to do this, but I haven't found any equivalent for proc NLIN. Would anyone have any ideas?
Here is the code I have so far: it outputs a dataset called output_data which contains my parameters: Asat, fi, th, and Rd. Running this will give me the parameter estimates and their standard errors in a separate table; I need the standard errors associated with each parameter to be included in the output_data dataset.
proc NLIN data=AQ method=Marquardt outest=status noitprint; where curve=4;
by curve;
f = (Asat+PPFD*fi-SQRT((fi*PPFD+Asat)*(fi*PPFD+Asat)-4*fi*Asat*th*PPFD))/(2*th)-Rd;
bounds -3<Rd<50;
parameters Asat= 5 to 60 by 1
fi= 0.2 to 0.8 by 0.1
Rd= 1E-8 to 10.1 by 0.5
th= 0 to 1 by 0.2;
model A=f;
output out=output_data parms=Asat fi th Rd predicted=Ahat student=student;
run;