[R] Looking for package to solve for exponent using newton's method

Prof J C Nash (U30A) nashjc at uottawa.ca
Fri Oct 11 17:11:53 CEST 2013


And if you need some extra digits:

require(Rmpfr)
testfn<-function(x){2^x+3^x-13}
myint<-c(mpfr(-5,precBits=1000),mpfr(5,precBits=1000))
myroot<-unirootR(testfn, myint, tol=1e-30)
myroot

John Nash

On 13-10-11 06:00 AM, r-help-request at r-project.org wrote:
> Message: 33
> Date: Thu, 10 Oct 2013 21:03:00 +0200
> From: Berend Hasselman<bhh at xs4all.nl>
> To: Ken Takagi<katakagi at bu.edu>
> Cc:r-help at stat.math.ethz.ch
> Subject: Re: [R] Looking for package to solve for exponent using
> 	newton's	method
> Message-ID:<A12B1F57-A38F-470D-A6D7-479F94125BC0 at xs4all.nl>
> Content-Type: text/plain; charset="us-ascii"
>
>
> On 10-10-2013, at 20:39, Ken Takagi<katakagi at bu.edu>  wrote:
>
>> >Hi,
>> >I'm looking for an R function/package that will let me solve problems of the
>> >type:
>> >
>> >13 = 2^x + 3^x.
>> >
>> >The answer to this example is x = 2, but I'm looking for solutions when x
>> >isn't so easily determined. Looking around, it seems that there is no
>> >algebraic solution for x, unless I'm mistaken.  Does anyone know a good
>> >package to solve these types of problems? Are there built in functions to do
>> >this?
>> >
> Univariate equations can be solved with uniroot, available in base R.
>
> You can also use package nleqslv for this but that is intended for systems of nonlinear equations.
> It does however solve your equation.
> There is also BB which is especially intended for large sparse systems.
>
> Berend
>



More information about the R-help mailing list