Time Series, Hidden Variables and Spatio-Temporal Ordinality Networks

被引:0
作者
Sudharsan Thiruvengadam
Jei Shian Tan
Karol Miller
机构
[1] University of Western Australia,
[2] BlueStem Pty. Ltd.,undefined
来源
Advances in Applied Clifford Algebras | 2020年 / 30卷
关键词
Time series; Conformal Geometric Algebra; Networks; Multi-variate systems; Forecasting;
D O I
暂无
中图分类号
学科分类号
摘要
In this work, a novel methodology for the modelling and forecasting of time series using higher-dimensional networks in R4,1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathbf{R }^{4,1}$$\end{document} space is presented. Time series data is partitioned, transformed and mapped into five-dimensional conformal space as a network which we call the ‘Spatio-Temporal Ordinality Network’ (STON). These STONs are characterised using specific Clifford Algebraic multivector functions which are found to be highly effective as featurised variables that forecast future states of the time series. The proposed STONs present as unique mathematical constructions that capture the algebra-geometric inter-dependencies in the historical behaviour of a time series system, thereby presenting governing expressions for forecasting based on historical values. Case studies on the seasonally adjusted Australian unemployment rate and the NASA Goddard Institute for Space Studies (GISS) Surface Temperature Analysis for Global Land-Ocean Temperature Index are furnished in this work and are compared against multilayer perceptrons (MLP), long short term memory (LSTM) neural networks, ARIMA and Holt-Winters methods. This formulation presents an alternative and effective modelling and forecasting paradigm for time series and multivariate systems with known or hidden variables.
引用
收藏
相关论文
共 118 条
  • [1] Pescetti D(1989)Mathematical modelling of hysteresis Il Nuovo Cimento D 11 1191-1216
  • [2] Chatfield C(1978)The Holtwinters forecasting procedure J. R. Stat. Soc. Ser. C (Appl. Stat.) 27 264-279
  • [3] Brunton BW(2016)Extracting spatial-temporal coherent patterns in large-scale neural recordings using dynamic mode decomposition J. Neurosci. Methods 258 1-15
  • [4] Johnson LA(2015)Discovering dynamic patterns from infectious disease data using dynamic mode decomposition Int. Health 7 139-145
  • [5] Ojemann JG(2019)Randomized dynamic mode decomposition SIAM J. Appl. Dyn. Syst. 18 1867-1891
  • [6] Kutz JN(2018)Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM Energy 148 461-468
  • [7] Proctor JL(2018)A deep learning model for short-term power load and probability density forecasting Energy 160 1186-1200
  • [8] Eckhoff PA(2019)Accurate photovoltaic power forecasting models using deep LSTM-RNN Neural Comput. Appl. 31 2727-2740
  • [9] Erichson NB(2019)A comparative study of series arima/mlp hybrid models for stock price forecasting Commun. Stat. Simul. Comput. 48 2625-2640
  • [10] Mathelin L(2007)Theory and applications of Clifford support vector machines J. Math. Imaging Vis. 28 29-46