Hi Members,
While fitting model through PROC NLMIXED, I am getting Negative value of
-2log-likelihoo. So far I knew that -2log Likelihood takes only positive
values as Likelihood varies between 0 to 1 and as a result
Log-likelihood varies between (- infinite, 0) and the value of - 2log
likelihood is always positive value.
data calib;
input Conc B3 ;
cards;
25.42 1.886
12.71 1.103
6.35 0.622
3.18 0.347
1.59 0.173
0.79 0.090
0.40 0.052
;
run;
proc nlmixed data=calib tech =newrhap;
parms top=5 bottom=0.01 EC50=50 hill=0.95 var=0.0001 ;
pred = bottom + (top-bottom)/(1+(EC50/conc)**hill);
bounds 0<bottom<0.05;
bounds top<10;
model B3~Normal(pred,(pred**2)*var );
run;
And I am getting following output:
Fit Statistics
-2 Log Likelihood -58.5
AIC (smaller is better) -48.5
AICC (smaller is better) -41.0
BIC (smaller is better) -45.3
Kindly let me know where I am going wrong.
--------------
Madan Gopal Kundu
Biostatistician, Ranbaxy Labs. Ltd.
Gurgaon, Haryana
India
Web: http://www.freewebs.com/madanstata
mobile: 91-9868788406
e-mail: [log in to unmask]
Best Jokes, Best Friends, Best Food and more. Go to http://in.promos.yahoo.com/groups/bestofyahoo/
|