Continuous and discrete best polynomial degree reduction with Jacobi and Hahn weights (Q281549)

Polynomial degree reduction (i.e., finding lower order polynomials of a given maximal order) algorithms are considered in this article and the degree reductions are made with respect to the best Jacobi least squares approximations. This means that the square-integrals of the norm are weighted with weights \(x^\alpha(1-x)^\beta\), the integral being taken over the unit interval. Among other things it is shown that the solution of the best approximation problem with respect to the aforementioned Jacobi norm is equivalent to the solution of the best approximation problem with respect to the norm computed by the weighted least squares norm of the Bézier coefficients.