[Date Index]
[Thread Index]
[Author Index]
Re: Selecting Real Roots, Again
*To*: mathgroup at smc.vnet.net
*Subject*: [mg67134] Re: Selecting Real Roots, Again
*From*: Peter Pein <petsie at dordos.net>
*Date*: Sat, 10 Jun 2006 04:53:31 -0400 (EDT)
*References*: <e6b5q5$d3c$1@smc.vnet.net>
*Sender*: owner-wri-mathgroup at wolfram.com
DOD schrieb:
> I've read the many posts already here about how to get mathematica to
> select real roots for you, but I have a slightly(very slighty, I
> thought) different problem ,and I don't know how to get mathematica to
> do what I want.
>
> I want to get the solution for a polynomial of the following form:
> d x^n + (1-d) x^2 =y
>
> so for example, I do
> Solve[.9 x^10 + .1 x^2 ==y,x]
> and I get a whole bunch of solution, very good. For my purposes, y
> lives in the [0,1], as must the solution. So I can see, by hand, which
> root I want; exactly one root is both real, and has solutions in my
> inverval. So I want to tell mathematica to:
>
> A: look at only solutions x* that are real over y in [0,1]
>
> and
>
> B: of those solutions, give the one x* that itself lies is [0,1].
>
> So, when I try to do something from reading previous posts, I cannot
> get it to work:
> In[24]:=
> Select[Solve[.9 x^10 + .1 x^2 ==y,x],Element[x/.#,Reals]&]
>
> Out[24]=
> {}
> or perhaps
> In[41]:=
> Select[Solve[.9 x^10 + .1 x^2
> ==y,x],Assuming[Element[y,Reals],Eement[x/.#,Reals]]&]
> Out[41]=
> {}
>
>
> So How to I tell mathematica to do this?
>
Hi,
you might try Reduce with assumptions:
In[1]:=
$Assumptions={0?x,x?1,0?y,y?1};
x0=x/.{ToRules[Refine[Reduce[
9/10 x^10+1/10x^2==y&&And@@$Assumptions,x]]]}
-->
{Root[-10*y + #1^2 + 9*#1^10 & , 2]}
You have to take the global assumptions into the call to Reduce[], since this function obviously doesn't care about $Assumptions (in ver. 5.1).
The addtitional Refine[] removes 0<=y<=1 from the output of Reduce.
hth,
Peter
Prev by Date:
**Re: Selecting Real Roots, Again**
Next by Date:
**Re: Selecting Real Roots, Again**
Previous by thread:
**Re: Selecting Real Roots, Again**
Next by thread:
**Re: Selecting Real Roots, Again**
| |