Nonparametric tests of independence for pairs of paths of stochastic processes

Abstract

Philip A. ErnstNonparametric tests of independence for pairs of paths of stochastic processes1. AbstractWe aim to build the first demonstrably correct statistical tests for testing independence (or dependence) of pairs of paths of time-dependent stochastic processes. Given two sequences of time-dependent observations Xjs and Yjs, we propose tests for the hypothesisH0 : (Xj) and (Yj) are independent, that are asymptotically exact and powerful under minimal assumptions.The relevant notions of dependence which we will consider are manifold: from whether pairs of paths of stochastic processes are independent, to applied consequences as one considers questions of attribution of factors for real-world phenomena, particularly relating to weather and climate. Important applied goals include an investigation of climate-related risks, such as sea-level rise and extreme weather events, particularly in the North Atlantic Ocean, how they correlate dynamically over medium and long terms, and how heavy-tailed they are. These goals could help with the estimation of costs of U.S. Naval vessels and U.S. Naval installations which may be at risk from extreme weather events.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 05, 2021
Source ID
N000142112672

Entities

People

  • Philip Ernst

Organizations

  • Office of Naval Research
  • Rice University
  • United States Navy

Tags

Fields of Study

  • Mathematics

Readers

  • Applied Combinatorial Optimization and Logic Circuit Design.
  • Statistical inference.
  • Theoretical Analysis.