[R] optimizing with nonlinear constraints
Rainer M Krug
Rainer at krugs.de
Wed Sep 30 10:46:59 CEST 2015
Hi
I have posted the following question on stackoverflow [1] but haven't
received a response yet. Has somebody here any idea, how to formulate
nonlinear constraints using e.g. the packages alabama or nlopr? The
question is pasted below.
I am really stuck with this.
Thanks,
Rainer
,
 I am successful in using constrOptim() for linear constraints (thanks to
 a comment in my earlier question), but I now have a problem with
 nonlinear constraints.

 Consider the following example:

 I want to fit a function FUN(X1, X2, ... X12)

 X1 = dep.a
 X2 = dep.b
 X3 = dep.c
 ... for all z0. , na. and zjoint.

 and have the following constraint
 (link to formulas:
 http://stackoverflow.com/questions/32845169/howtooptimizewithnonlinearconstraintsinr
 )
 
 With some reformating and h > 0 and LAI > 0 I get for thew first one

 X1 <= LAI^X2 / X3 < 1  X1

 where LAI is simply a number which constant for each fit.

 I saw the packages alabama and nloptr but I don't manage to get my head
 around how I can specify these types of constraints?

 Also a web search did not yield any helpful information for me.

 Could somebody provide some info how I can convert these constraints
 into arguments for the fitting functions (e.g. hin, heq in the alabama
 package)?
`
Footnotes:
[1] http://stackoverflow.com/questions/32845169/howtooptimizewithnonlinearconstraintsinr

Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany)
Centre of Excellence for Invasion Biology
Stellenbosch University
South Africa
Tel : +33  (0)9 53 10 27 44
Cell: +33  (0)6 85 62 59 98
Fax : +33  (0)9 58 10 27 44
Fax (D): +49  (0)3 21 21 25 22 44
email: Rainer at krugs.de
Skype: RMkrug
PGP: 0x0F52F982
 next part 
A nontext attachment was scrubbed...
Name: not available
Type: application/pgpsignature
Size: 454 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/rhelp/attachments/20150930/a30cef3d/attachment.bin>
More information about the Rhelp
mailing list