On Numerical Estimation of Joint Probability Distribution from Lebesgue Integral Quadratures

9 Pages Posted: 24 Aug 2018 Last revised: 1 Dec 2020

Date Written: July 19, 2018

Abstract

An important application of Lebesgue integral quadrature is developed. Given two random processes, 𝒇(x) and 𝒈(x), two generalized eigenvalue problems can be formulated and solved. In addition to obtaining two Lebesgue quadratures (for 𝒇 and 𝒈) from two eigenproblems, the projections of 𝒇- and 𝒈- eigenvectors on each other allow to build a joint distribution estimator, the most general form of which is a density-matrix correlation. Examples of the density-matrix correlation can be a value-correlation V(𝒇;𝒈), similar to a regular correlation concept, and a new one, a probability-correlation P(𝒇;𝒈). If Christoffel function average is used instead of regular average the approach can be extended to an estimation of joint probability of three and more random processes. The theory is implemented numerically; the software is available under the GPLv3 license.

Keywords: Machine Learning, Quality Criteria, Non-Gaussian, Lebesgue

JEL Classification: C63

Suggested Citation

Malyshkin, Vladislav, On Numerical Estimation of Joint Probability Distribution from Lebesgue Integral Quadratures (July 19, 2018). Available at SSRN: https://ssrn.com/abstract=3229347 or http://dx.doi.org/10.2139/ssrn.3229347

Vladislav Malyshkin (Contact Author)

Ioffe Institute ( email )

Politekhnicheskaya 26
St Petersburg, 194021
Russia

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
46
Abstract Views
554
PlumX Metrics