Discovering Multiple Phases of Dynamics by Dissecting Multivariate Time Series

03/08/2021
by   Xiaodong Wang, et al.
0

We proposed a data-driven approach to dissect multivariate time series in order to discover multiple phases underlying dynamics of complex systems. This computing approach is developed as a multiple-dimension version of Hierarchical Factor Segmentation(HFS) technique. This expanded approach proposes a systematic protocol of choosing various extreme events in multi-dimensional space. Upon each chosen event, an empirical distribution of event-recurrence, or waiting time between the excursions, is fitted by a geometric distribution with time-varying parameters. Iterative fittings are performed across all chosen events. We then collect and summarize the local recurrent patterns into a global dynamic mechanism. Clustering is applied for partitioning the whole time period into alternating segments, in which variables are identically distributed. Feature weighting techniques are also considered to compensate for some drawbacks of clustering. Our simulation results show that this expanded approach can even detect systematic differences when the joint distribution varies. In real data experiments, we analyze the relationship from returns, trading volume, and transaction number of a single, as well as of multiple stocks in S P500. We can successfully not only map out volatile periods but also provide potential associative links between stocks.

READ FULL TEXT

page 15

page 18

page 19

page 21

page 27

page 28

page 29

page 30

research
11/28/2019

Analysis of Hydrological and Suspended Sediment Events from Mad River Wastershed using Multivariate Time Series Clustering

Hydrological storm events are a primary driver for transporting water qu...
research
10/27/2018

Time series clustering based on the characterisation of segment typologies

Time series clustering is the process of grouping time series with respe...
research
05/06/2015

Cats & Co: Categorical Time Series Coclustering

We suggest a novel method of clustering and exploratory analysis of temp...
research
01/23/2021

Unraveling S P500 stock volatility and networks – An encoding and decoding approach

We extend the Hierarchical Factor Segmentation(HFS) algorithm for discov...
research
05/03/2016

Temporal Clustering of Time Series via Threshold Autoregressive Models: Application to Commodity Prices

This study aimed to find temporal clusters for several commodity prices ...
research
01/22/2019

Unsupervised Automated Event Detection using an Iterative Clustering based Segmentation Approach

A class of vision problems, less commonly studied, consists of detecting...
research
08/29/2016

Discovering Patterns in Time-Varying Graphs: A Triclustering Approach

This paper introduces a novel technique to track structures in time vary...

Please sign up or login with your details

Forgot password? Click here to reset