On the approximation of the probability density function of the randomized non-autonomous complete linear differential equation

TL;DR

This paper develops methods to approximate the probability density function of solutions to randomized non-autonomous linear differential equations with stochastic coefficients and initial conditions, using transformations and expansions.

Contribution

It introduces a novel approach combining Random Variable Transformation and Karhunen-Loeve expansions to approximate the solution's probability density function.

Findings

Effective approximation of the solution's PDF demonstrated through numerical experiments.

Applicable to a wide range of stochastic differential equations.

Provides a framework for analyzing stochastic processes in linear differential equations.

Abstract

In this paper we study the randomized non-autonomous complete linear differential equation. The diffusion coefficient and the source term in the differential equation are assumed to be stochastic processes and the initial condition is treated as a random variable on an underlying complete probability space. The solution to this random differential equation is a stochastic process. Any stochastic process is determined by its finite-dimensional joint distributions. In this paper, the main goal is to obtain the probability density function of the solution process (the first finite-dimensional distribution) under mild conditions. The solution process is expressed by means of Lebesgue integrals of the data stochastic processes, which, in general, cannot be computed in an exact manner, therefore approximations for its probability density function are constructed. The key tools applied to construct the approximations are the Random Variable Transformation technique and Karhunen-Loeve expansions. Our results can be applied to a large variety of examples. Finally, several numerical experiments illustrate the potentiality of our findings.