Can I read Drift Estimation in Stochastic Flows Using Kernel Integral Operators on EtoBox?

Drift Estimation in Stochastic Flows Using Kernel Integral Operators by Das, Suddhasattwa is a scholarly article available to read on EtoBox.

What is Drift Estimation in Stochastic Flows Using Kernel Integral Operators about?

A stochastic differential equation (SDE) describes a motion in which a particle is governed simultaneously by the direction provided by a vector field / drift, and the scattering effects of white noise. The resulting motion can only be described as a random process instead of a solution curve. Due to the non-deterministic nature of this motion, the task of determining the drift from data is quite challenging, since the data does not directly represent the directional information of the flow. This paper describes an interpretation of vector field as a conditional expectation, which makes its estimation feasible via kernel-integral methods. It presents a numerical procedure based on kernel integral operators, that computes this expectation. In addition, some techniques are presented which can overcome the challenge of dimensionality if the SDE's carry some structure enabling sparsity.

Author

Das, Suddhasattwa

Published

2024

Language

EN