Re: [R] problem in fitting model in NLS function
Bad scaling will waste a lot of everyone's time.
I put the data in a data frame mdat, then
library(nlmrt)
mdat<-read.csv("muzzamil.csv", header=T)
fmn <- nlxb(y~a * (x^b), data=mdat, start=c(a=1,b=1), trace=T)
fm <- nls(y~a * (x^b), data=mdat, start=c(a=1,b=1), trace=T)
fmn2 <- nlxb(y~a2 * ((x-1979)^b2), data=mdat, start=c(a2=1,b2=1), trace=T)
fm2 <- nls(y~a2 * ((x-1979)^b2), data=mdat, start=c(a2=1,b2=1), trace=T)
found a possible solution quickly with the last 2 forms. Without
adjusting the x data, you get a very ill-conditioned problem.
JN
Message: 21
Date: Wed, 14 Nov 2012 03:04:56 -0800 (PST)
From: muzammil786
To: r-help@r-project.org
Subject: Re: [R] problem in fitting model in NLS function
Message-ID: <1352891096449-4649473.p...@n4.nabble.com>
Content-Type: text/plain; charset=us-ascii
Dear David,
I am having a similar problem on this data (given below). I have tried
several starting values for a and b, but it is consistently giving me:
*fm <- nls(y~f(x,a,b), data.frame(x,y), start=c(a=1,b=1))
Error in numericDeriv(form[[3L]], names(ind), env) :
Missing value or an infinity produced when evaluating the model*
The function f(x,a,b) is defined here:
*f <- function(x,a,b) {a * I(x^b)} *
The data is given here:
* x y
1 1980 1
2 1981 1
3 1982 1
4 1983 3
5 1984 3
6 1985 5
7 1986 8
8 1987 8
9 1988 9
10 1989 12
11 1990 15
12 1991 24
13 1992 33
14 1993 44
15 1994 62
16 1995 68
17 1996 81
18 1997 87
19 1998 102
20 1999 114
21 2000 123
22 2001 135
23 2002 144
24 2003 158
25 2004 172
26 2005 188
27 2006 197
28 2007 224
29 2008 234
30 2009 254
31 2010 278
32 2011 312
33 2012 317*
I shall be grateful if you could guide me what's wrong here.
Cheers.
Muzammil
--
View this message in context:
http://r.789695.n4.nabble.com/problem-in-fitting-model-in-NLS-function-tp4345082p4649473.html
Sent from the R help mailing list archive at Nabble.com.
__
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.
[R] vectorized uni-root?
The package rootoned on http://r-forge.r-project.org/R/?group_id=395 has an all-R version of zeroin (the algorithm of uniroot). This should also be in Rmpfr by Martin M., as it was set up for that use. I suspect it can be vectorized fairly easily. However, it may be simpler to write, or else abstract from that code, a vectorized false position or secant code (same formula, FP must have end points with opposite sign function values, so is `safer'). This avoids the bisection step and simplifies things, but may be a bit less efficient sometimes. JN Date: Fri, 9 Nov 2012 14:58:55 + From: Ravi Varadhan To: "'ivo.we...@gmail.com'" Cc: "r-help@r-project.org" Subject: Re: [R] vectorized uni-root? Message-ID: <2f9ea67ef9ae1c48a147cb41be2e15c32e8...@dom-eb-mail1.win.ad.jhu.edu> Content-Type: text/plain Hi Ivo, The only problem is that uniroot() actually uses Brent's algorithm, and is based on the C code from netlib (there is also Fortran code available - zeroin.f). Brent's algorithm is more sophisticated than a simple bisection approach that you have vectorized. It combines bisection and secant methods along with some quadratic interpolation. The idea behind this hybrid approach (which by the way is a fundamental theme in all of numerical analysis) is to get faster convergence while not sacrificing the global convergence of bisection. So, the existing uniroot() will not be deprecated unless you can vectorize Brent's hybrid root-finding approach! Best, Ravi __ 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.
[R] Error message in nmkb()
I would guess -- we don't have a reproducible code -- that you have an undefined function value. Since nmkb is entirely in R to my knowledge, you could simply take the source code and modify it to display the elements in the test that cause the trouble. Then load the source rather than using library(dfoptim). Alternatively, add to your objective function some code to display the parameters before evaluation and the resulting value that is returned. Messy but pretty effective to reveal if and where the function is giving trouble. JN On 12-11-06 06:00 AM, r-help-requ...@r-project.org wrote: Date: Mon, 5 Nov 2012 01:56:30 -0800 (PST) From: MMar86 To:r-help@r-project.org Subject: [R] Error message in nmkb() Message-ID:<1352109390759-4648459.p...@n4.nabble.com> Content-Type: text/plain; charset=UTF-8 Hallo together, I am trying to use the nmkb() optimizer and I have problems using the function, as it causes the following error message Fehler (error)* in while (nf < maxfeval & restarts < restarts.max & dist > ftol & : Fehlender Wert (missing value)* , wo (where)* TRUE/FALSE n?tig ist (is required)* *translation Do I need to adjust the control ? my.optpar3<-nmkb(par=par.start,fn=my.loglike.normal,lower=constLo,upper=999, control=list(restarts.max=9)) If you need more info, please let me know. I would appreciate some help, thank you very much. Best, Marcial __ 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.
