next up previous contents
Next: Joyal bounds Up: Newton theorem Previous: Mathematical background   Contents

Implementation

This procedure is able to give an upper bound on the absolute value of the roots as soon as all the roots are real (this is checked using Huat theorem, section 5.5.4). The syntax is:
 
int Newton_Second_Bound_Interval(int Degree,VECTOR &Coeff,double *bound);
with: The procedure returns 0 if Degree=0 or Coeff(Degree+1)=0, 1 otherwise. We may also compute a lower bound on the absolute value of the root by using:
 
int Newton_Second_Bound_Inverse_Interval(int Degree,VECTOR &Coeff,double *bound);
An upper and lower bound on the absolute value of the roots may be compute by:
 
int Newton_Second_Bound_Interval(int Degree,VECTOR &Coeff,INTERVAL &Bound);
In that case if Bound=[a,b] we have for all root $x$: a$ \le
\vert x\vert \le$ b.

Similar procedures exist for interval polynomial:

 
int Newton_Second_Bound_Interval(int Degree,INTERVAL_VECTOR &Coeff,INTERVAL &Bound);
int Newton_Second_Bound_Inverse_Interval(int Degree,INTERVAL_VECTOR &Coeff,INTERVAL &Bound);
int Newton_Second_Bound_Interval(int Degree,INTERVAL_VECTOR &Coeff,INTERVAL &L, INTERVAL &U);
where


next up previous contents
Next: Joyal bounds Up: Newton theorem Previous: Mathematical background   Contents
Jean-Pierre Merlet 2012-12-20