Nonlinear Principal Components and Long-run Implications of Multivariate Diffusions

TL;DR

This paper introduces a method for extracting nonlinear principal components (NPCs) in multivariate diffusion processes, providing new interpretations, estimation techniques, and implications for long-term analysis of complex stochastic systems.

Contribution

It develops a general framework for NPCs with broad constraints, offers a novel interpretation via Markov diffusion theory, and proposes a sieve-based estimation method from discrete data.

Findings

NPCs maximize long-run variation under smoothness constraints

NPCs behave like scalar autoregressions with heteroskedasticity

Proposes a sieve method for estimating NPCs from sampled data

Abstract

We investigate a method for extracting nonlinear principal components (NPCs). These NPCs maximize variation subject to smoothness and orthogonality constraints; but we allow for a general class of constraints and multivariate probability densities, including densities without compact support and even densities with algebraic tails. We provide primitive sufficient conditions for the existence of these NPCs. By exploiting the theory of continuous-time, reversible Markov diffusion processes, we give a different interpretation of these NPCs and the smoothness constraints. When the diffusion matrix is used to enforce smoothness, the NPCs maximize long-run variation relative to the overall variation subject to orthogonality constraints. Moreover, the NPCs behave as scalar autoregressions with heteroskedastic innovations; this supports semiparametric identification and estimation of a multivariate reversible diffusion process and tests of the overidentifying restrictions implied by such a process from low frequency data. We also explore implications for stationary, possibly non-reversible diffusion processes. Finally, we suggest a sieve method to estimate the NPCs from discretely-sampled data.