Skip to Main Content

Proceedings of the American Mathematical Society

Published by the American Mathematical Society since 1950, Proceedings of the American Mathematical Society is devoted to shorter research articles in all areas of pure and applied mathematics.

ISSN 1088-6826 (online) ISSN 0002-9939 (print)

The 2020 MCQ for Proceedings of the American Mathematical Society is 0.85.

What is MCQ? The Mathematical Citation Quotient (MCQ) measures journal impact by looking at citations over a five-year period. Subscribers to MathSciNet may click through for more detailed information.

 

Revisiting two theorems of Curto and Fialkow on moment matrices
HTML articles powered by AMS MathViewer

by Monique Laurent PDF
Proc. Amer. Math. Soc. 133 (2005), 2965-2976 Request permission

Abstract:

We revisit two results of Curto and Fialkow on moment matrices. The first result asserts that every sequence $y\in \mathbb {R}^{\mathbb {Z}^n_+}$ whose moment matrix $M(y)$ is positive semidefinite and has finite rank $r$ is the sequence of moments of an $r$-atomic nonnegative measure $\mu$ on $\mathbb {R}^n$. We give an alternative proof for this result, using algebraic tools (the Nullstellensatz) in place of the functional analytic tools used in the original proof of Curto and Fialkow. An easy observation is the existence of interpolation polynomials at the atoms of the measure $\mu$ having degree at most $t$ if the principal submatrix $M_t(y)$ of $M(y)$ (indexed by all monomials of degree $\le t$) has full rank $r$. This observation enables us to shortcut the proof of the following result. Consider a basic closed semialgebraic set $F=\{x\in \mathbb {R}^n\mid h_1(x)\ge 0, \ldots ,h_m(x)\ge 0\}$, where $h_j\in \mathbb {R}[x_1,\ldots ,x_n]$ and $d:=\operatorname {max}_{j=1}^m \lceil \operatorname {deg}(h_j)/2\rceil$. If $M_t(y)$ is positive semidefinite and has a flat extension $M_{t+d}(y)$ such that all localizing matrices $M_{t}(h_j\ast y)$ are positive semidefinite, then $y$ has an atomic representing measure supported by $F$. We also review an application of this result to the problem of minimizing a polynomial over the set $F$.
References
Similar Articles
Additional Information
  • Monique Laurent
  • Affiliation: Centrum voor Wiskunde en Informatica, Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
  • Email: M.Laurent@cwi.nl
  • Received by editor(s): January 16, 2004
  • Published electronically: May 9, 2005
  • Additional Notes: This work was supported by the Netherlands Organisation for Scientific Research grant NWO 639.032.203
  • Communicated by: Lance W. Small
  • © Copyright 2005 American Mathematical Society
  • Journal: Proc. Amer. Math. Soc. 133 (2005), 2965-2976
  • MSC (2000): Primary 44A30, 13J30, 14P10, 90C22
  • DOI: https://doi.org/10.1090/S0002-9939-05-08133-5
  • MathSciNet review: 2159775