I tried to use calcMin with a function that uses a number of ...
arguments (all args from resid on) besides the vector of parameters
being fit. Same idea as optim, nlm, nlminb for which this form of ...
syntax works. But with calcMin I get an error regarding unused
arguments. No partial matches to previous arguments that I can see.
Anybody know the reason or fix for this?
calcMin(pvec=data.frame(val=par,min=rep(.000001,length(par)),max=rep(100
0000,length(par)),active=rep(TRUE,length(par))),func=optimwrap2,
resid=resid,caa=caa,na=na,vcode=vcode,maa=maa,ny=ny,nind=nind,qage=qage,
selmod=selmod,oldagei=oldagei,vpaflag=vpaflag)
Error in f(x, ...) :
unused argument(s) (resid = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0), caa = c(0.344919912565012,
1.36925481463155, 4.94519930946541, 5.12996667120799, 11.0171415280463,
9.4432641668968, 4.7216320834484, 3.14775472229893, 1.57387736114947,
3.14775472229893,
>
And an example using optim to address concerns about the user function
itself. [I'm trying to concoct a solver that handles far-off initial
parameter values without getting stuck in local minima. I'm searching
for something faster and more intelligent than this example.]
> for(i in 1:9) {
+
fitpars=optim(par=parset1,fn=optimwrap,control=list(maxit=100),resid=res
id,caa=caa,na=na,vcode=vcode,maa=maa,ny=ny,nind=nind,qage=qage,selmod=se
lmod,oldagei=oldagei,vpaflag=vpaflag)
+ parset2=fitpars$par
+ parval2=fitpars$value
+ print(parval2)
+ print(parset2)
+ if((parval1-parval2)>1.0) {
+ parval1=parval2
+ parset1=parset2
+ }
+ if((parval1-parval2)<1.0 & parval2>0.1) {
+ parset1=par*pardown[i]
+ }
+ }
[1] 193.1500
[1] 43.84105 43.47576 42.91166 41.81009 42.57930
[1] 15.69411
[1] 13.35845 12.52650 12.28231 11.78171 11.29699
[1] 15.65625
[1] 11.91989 11.10244 10.84155 10.34886 9.85170
[1] 15.45980
[1] 10.374731 9.569294 9.302000 8.779963 8.325952
[1] 14.85332
[1] 9.105026 8.293079 8.004561 7.521170 7.080387
[1] 8.18217
[1] 6.209737 5.876993 5.229392 4.887182 4.102208
[1] 6.720143
[1] 6.180562 5.276061 5.057946 4.561191 4.317552
[1] 0.04003793
[1] 3.754422 2.871289 2.876904 2.091451 1.806243
[1] 0.04003793
[1] 3.754422 2.871289 2.876904 2.091451 1.806243
Mark Fowler
Population Ecology Division
Bedford Inst of Oceanography
Dept Fisheries & Oceans
Dartmouth NS Canada
B2Y 4A2
Tel. (902) 426-3529
Fax (902) 426-9710
Email mark.fow...@dfo-mpo.gc.ca
Home Tel. (902) 461-0708
Home Email mark.fow...@ns.sympatico.ca
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
