The nearest polynomial of lower degree

Suppose one is working with the polynomial p(x) = −0.99B3 0(x)−1.03B 1(x)− 0.33B 2(x) + B 3 3(x) expressed in terms of degree 3 Bernstein polynomials and suppose one has reason to believe that p(x) is really of degree 2. Applying the degree-reducing procedure [6] one gets −0.99B2 0(x)− 1.05B 1(x) + B 2(x). But is this correct, or have we treated p(x) in a Procrustean 1 fashion? Checking by converting p(x) to the power basis, we find that the coefficient of x is 0.11. Is this zero or not?

[1]  Robert M. Corless,et al.  Pseudospectra of Matrix Polynomials that Are Expressed in Alternative Bases , 2007, Math. Comput. Sci..

[2]  A. Shakoori.  The Bézout matrix in the Lagrange basis , 2004 .

[3]  Robert M. Corless,et al.  The nearest polynomial with a given zero, revisited , 2005, SIGS.

[4]  Robert M Corless,et al.  Bernstein Bases are Optimal , but , sometimes , Lagrange Bases are Better , 2004 .

[5]  R. M. Corless,et al.  Generalized Companion Matrices in the Lagrange Bases , 2004 .

[6]  Guido Brunnett,et al.  The geometry of optimal degree reduction of Bézier curves , 1996, Comput. Aided Geom. Des..

[7]  Lloyd N. Trefethen,et al.  Barycentric Lagrange Interpolation , 2004, SIAM Rev..

[8]  Rida T. Farouki,et al.  Algorithms for polynomials in Bernstein form , 1988, Comput. Aided Geom. Des..

[9]  A. Amiraslani.  Dividing polynomials when you only know their values , 2004 .

[10]  Hans J. Stetter,et al.  The nearest polynomial with a given zero, and similar problems , 1999, SIGS.

[11]  Rida T. Farouki,et al.  On the numerical condition of polynomials in Bernstein form , 1987, Comput. Aided Geom. Des..

[12]  Young Joon Ahn,et al.  Constrained polynomial degree reduction in the L2-norm equals best weighted Euclidean approximation of Bézier coefficients , 2004, Comput. Aided Geom. Des..

[13]  Robert M. Corless,et al.  Polyno-mial Algebra by Values , 2004 .

[14]  Thomas Hermann.  On the stability of polynomial transformations between Taylor, Bernstein and Hermite forms , 2005, Numerical Algorithms.

[15]  N. Higham.  The numerical stability of barycentric Lagrange interpolation , 2004 .

[16]  Erich Kaltofen,et al.  Efficient algorithms for computing the nearest polynomial with constrained roots , 1998, ISSAC '98.

[17]  T. J. Rivlin.  Chebyshev polynomials : from approximation theory to algebra and number theory , 1990 .