non linear regression

e3flator — Tue, 07 Aug 2012 09:18:01 GMT

Hi, I've to approximate some data using non linear regression. Both PROC MODEL and PROC NLIN do "non linear regression" which one is better?

I've tried both on some sample data. The PROC NLIN gives results, instead the PROC MODEL gives this error:

ERROR: The Newton method Jacobian matrix of partial derivatives of the equations with respect to

the variables to be solved is singular at observation 1, for iteration 2. The system of

equations cannot be solved.

data vector_def;

x=1; y= 0.00; output;

x=2; y= 0.00; output;

x=3; y= 0.00; output;

x=4; y= 0.05; output;

x=5; y= 0.06; output;

x=6; y= 0.07; output;

x=7; y= 0.06; output;

x=8; y= 0.15; output;

x=9; y= 0.18; output;

x=10; y= 0.30; output;

x=11; y= 0.71; output;

x=12; y= 0.80; output;

x=13; y= 1.83; output;

x=14; y= 2.51; output;

x=15; y= 3.99; output;

x=16; y= 7.58; output;

x=17; y= 24.32; output;

run;

proc nlin data=vector_def method=newton ;

parms b0=1 b1=0.2 b2=0 to 0.4 by 0.1;

model y=b0*exp(b1*x)+b2;

output out=fitexp p=yfit r=resid;

run;

quit;

proc model data=vector_def outparms=outparms;

endogenous x;

exogenous y;

y = b0 * exp(b1*x) + b2;

fit y;

solve / out=vector_def_out ;

run;

quit;

Is there a macro variable to intercept the proc model error?

Thanks in advance

Federica

Re: non linear regression

Reeza — Tue, 07 Aug 2012 16:16:11 GMT

Do you have time series data?

If so then you need to work with Model, if not then you can try working with NLIN and NLMIXED

Re: non linear regression

SteveDenham — Wed, 08 Aug 2012 12:39:34 GMT

Federica,

You have confused the endogenous and exogenous variables in PROC MODEL. The independent variable x should be in the exogenous statement and the dependent variable y in the endogenous statement.

Good luck.

Steve Denham

topic non linear regression in Statistical Procedures

non linear regression

Re: non linear regression

Re: non linear regression