R help - May 2006 - problems with optimize (again)

Can someone please explain what the $minimum result of the optimize
function actually is?

I'm trying to optimize the function:
fitIT<-function(ampFac,ts_wave1,ts_template){
    template<-stretchWaveTime(ts_template,ampFac)
    fit<-calcFit(ts_wave1,template)
    return(fit)
}

with
>optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1])
$minimum
[1] 3.764685

$objective
[1] 1.037864

however when I run
> calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864)) 
[1] 0.3059954

I thought $minimum and the result of the calcFit function should be the
same.

As a further question I'm going to need to nest optimize functions so I
can optimize over two variables. How can I get the minimum and objective
back from the nested function f, as well as the value to be minimised?

I hope this makes sense and I apologise for the lack of reproducible
code. If it will help I can include that in a further e-mail.
Thanks
Tom

you should use:

optmz <- optimize(fitIT, interval = c(0.5, 4),
       ts_wave1 = test.data[, 1], ts_template = test.data[, 1])

calcFit(test.data[, 1], stretchWaveTime(test.data[, 1], optmz$minimum)

which should give the same as optmz$objective, i.e., the optmz$minimum 
is the value of 'ampFac' argument for which fitIT() attains its 
minimum value; the value of fitIT() at optmz$minimum is 
optmz$objective.

I hope it helps.

Best,
Dimitris

----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://www.med.kuleuven.be/biostat/
     http://www.student.kuleuven.be/~m0390867/dimitris.htm



----- Original Message ----- 
From: "tom wright" <tom at maladmin.com>
To: "R-Stat Help" <R-help at stat.math.ethz.ch>
Sent: Wednesday, May 10, 2006 2:18 PM
Subject: [R] problems with optimize (again)

> Can someone please explain what the $minimum result of the optimize
> function actually is?
>
> I'm trying to optimize the function:
> fitIT<-function(ampFac,ts_wave1,ts_template){
>    template<-stretchWaveTime(ts_template,ampFac)
>    fit<-calcFit(ts_wave1,template)
>    return(fit)
> }
>
> with
>>optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1])
> $minimum
> [1] 3.764685
>
> $objective
> [1] 1.037864
>
> however when I run
>> calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864))
> [1] 0.3059954
>
> I thought $minimum and the result of the calcFit function should be 
> the
> same.
>
> As a further question I'm going to need to nest optimize functions 
> so I
> can optimize over two variables. How can I get the minimum and 
> objective
> back from the nested function f, as well as the value to be 
> minimised?
>
> I hope this makes sense and I apologise for the lack of reproducible
> code. If it will help I can include that in a further e-mail.
> Thanks
> Tom
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
> 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

tom wright wrote:
> Can someone please explain what the $minimum result of the optimize
> function actually is?
> 
> I'm trying to optimize the function:
> fitIT<-function(ampFac,ts_wave1,ts_template){
>     template<-stretchWaveTime(ts_template,ampFac)
>     fit<-calcFit(ts_wave1,template)
>     return(fit)
> }
> 
> with
> 
>>optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1])
> 
> $minimum
> [1] 3.764685
> 
> $objective
> [1] 1.037864
> 
> however when I run
> 
>>calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864)) 
> 
> [1] 0.3059954

What is calcFit?
> I thought $minimum and the result of the calcFit function should be the
> same.
Why should to different algorithms (at least I guess so, since you don't 
say what calcFit() does) find the same *local* minimum?

> As a further question I'm going to need to nest optimize functions so I
> can optimize over two variables. How can I get the minimum and objective
> back from the nested function f, as well as the value to be minimised?

Use optim().

Uwe Ligges

> I hope this makes sense and I apologise for the lack of reproducible
> code. If it will help I can include that in a further e-mail.
> Thanks
> Tom
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html