A detection analysis for temporal memory patterns at different time-scales

TL;DR

This paper presents a new statistical methodology using latency analysis to detect memory dependence patterns in event sequences across multiple time scales, with applications in economics.

Contribution

It introduces a novel latency-based statistical test for uncovering time-scale dependence patterns and assesses renewal properties in time-series data.

Findings

Detects memory dependence across diverse time scales

Evaluates renewal assumption through aging experiments

Provides graphical tools for dependence pattern visualization

Abstract

This paper introduces a novel methodology that utilizes latency to unveil time-series dependence patterns. A customized statistical test detects memory dependence in event sequences by analyzing their inter-event time distributions. Synthetic experiments based on the renewal-aging property assess the impact of observer latency on the renewal property. Our test uncovers memory patterns across diverse time scales, emphasizing the event sequence's probability structure beyond correlations. The time series analysis produces a statistical test and graphical plots which helps to detect dependence patterns among events at different time-scales if any. Furthermore, the test evaluates the renewal assumption through aging experiments, offering valuable applications in time-series analysis within economics.