Hi,
In your example, the Nelder-Mead method is a lot faster and finds a lower
minimum to your function than the SANN method.
I also tried to look into your kalman filter, but I got lost. The notation you
use is hard to decipher.
Neuro
#No optimization
> kalman(startwerte*0.9)[1] -265.5831
#Optim with Nelder-Mead>
optim(startwerte*0.9,kalman,method="Nelder-Mead")$par[1] 0.072320797
0.934972315 0.004593847 0.001847743
$value[1] -305.8315
$countsfunction gradient 255 NA
$convergence[1] 0
$messageNULL
#Optim with SANN
> optim(startwerte*0.9,kalman,method="SANN") $par[1] 0.003154530
> 1.009002248 -0.005645733 0.002603398
$value[1] -294.2174
$countsfunction gradient 10000 NA
$convergence[1] 0
$messageNULL
_________________________________________________________________
Soyez parmi les premiers à essayer Windows Live Mail.
4911fb2b2e6d
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
