Reciprocal first-order second-moment method

TL;DR

This paper introduces a reciprocal parameter substitution technique that enhances the accuracy of first-order probabilistic analysis with minimal additional computational effort by leveraging the reciprocal relation of objective functions and parameters.

Contribution

It presents a novel parameter substitution method that improves first-order probabilistic analysis accuracy using reciprocal relations, with a clear transformation process for different distributions.

Findings

Significant accuracy improvement in probabilistic analysis

Minimal additional computational cost

Applicable to various distribution cases

Abstract

This paper shows a simple parameter substitution, which makes use of the reciprocal relation of typical objective functions with typical random parameters. Thereby, the accuracy of first-order probabilistic analysis improves significantly at almost no additional computational cost. The parameter substitution requires a transformation of the stochastic distribution of the substituted parameter, which is explained for different cases.