A new time-series model based on quantum walk

TL;DR

This paper introduces a novel time-series model based on quantum walks, capable of handling non-stationary data, unlike traditional models like ARMA or GARCH, which require stationarity.

Contribution

The paper presents a new quantum walk-based time-series model that does not assume stationarity, expanding the applicability to non-stationary data.

Findings

Model effectively handles non-stationary time series

Quantum walk properties enable new analysis methods

Potential for improved financial and economic data modeling

Abstract

The quantum walk (QW) was introduced as a quantum counterpart of the classical random walk. A number of non-classical properties of the QW have been shown, e.g., ballistic spreading, anti-bellshaped limit density, localization. Since around 2000, extensive research has been conducted in both theoretical aspects as well as the practical application of QWs. However, the application of a QW to the time-series analysis is not known. On the other hand, it is well known that the ARMA or GARCH models have been widely used in economics and finance. These models are studied under some suitable stationarity conditions. In this paper, we propose a new time-series model based on the QW, which does not assume such a stationarity. Therefore, our method would be applicable to the non-stationary time series.