Mean squared error minimization for inverse moment problems

From MaRDI portal




Abstract: We consider the problem of approximating the unknown density uinL2(Omega,lambda) of a measure mu on OmegasubsetRn, absolutely continuous with respect to some given reference measure lambda, from the only knowledge of finitely many moments of mu. Given dinN and moments of order d, we provide a polynomial pd which minimizes the mean square error int(up)2dlambda over all polynomials p of degree at most d. If there is no additional requirement, pd is obtained as solution of a linear system. In addition, if pd is expressed in the basis of polynomials that are orthonormal with respect to lambda, its vector of coefficients is just the vector of given moments and no computation is needed. Moreover pdou in L2(Omega,lambda) as doinfty. In general nonnegativity of pd is not guaranteed even though u is nonnegative. However, with this additional nonnegativity requirement one obtains analogous results but computing pdgeq0 that minimizes int(up)2dlambda now requires solving an appropriate semidefinite program. We have tested the approach on some applications arising from the reconstruction of geometrical objects and the approximation of solutions of nonlinear differential equations. In all cases our results are significantly better than those obtained with the maximum entropy technique for estimating u.



Cites work



Describes a project that uses

Uses Software





This page was built for publication: Mean squared error minimization for inverse moment problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q741143)