A New Machine Learning Forecasting Algorithm Based on Bivariate Copula Functions

A novel forecasting method based on copula functions is proposed. It consists of an iterative algorithm in which a dependent variable is decomposed as a sum of error terms, where each one of them is estimated identifying the input variable which best “copulate” with it. The method has been tested over popular reference datasets, achieving competitive results in comparison with other well-known machine learning techniques.

[1]  G. Venter.  TAILS OF COPULAS , 2013 .

[2]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[3]  J. Friedman.  Stochastic gradient boosting , 2002 .

[4]  M. Bhatti,et al.  Co-Movement, Dependence Structure and Ethical Investment Funds under GFC , 2019, Theoretical Economics Letters.

[5]  Lutz Prechelt,et al.  Early Stopping-But When? , 1996, Neural Networks: Tricks of the Trade.

[6]  Gregor N. F. Weiß.  Copula-GARCH versus dynamic conditional correlation: an empirical study on VaR and ES forecasting accuracy , 2013 .

[7]  W. Whitt.  Bivariate Distributions with Given Marginals , 1976 .

[8]  Pravin K. Trivedi,et al.  Using Trivariate Copulas to Model Sample Selection and Treatment Effects , 2006 .

[9]  A. Kresta,et al.  Application of GARCH-Copula Model in Portfolio Optimization , 2015 .

[10]  Bruce W. Schmeiser.  Random Variate Generation: A Survey. , 1980 .

[11]  Stoyan V. Stoyanov,et al.  Portfolio selection based on a simulated copula , 2010 .

[12]  Roger M. Cooke,et al.  Probability Density Decomposition for Conditionally Dependent Random Variables Modeled by Vines , 2001, Annals of Mathematics and Artificial Intelligence.

[13]  M. Haugh,et al.  An Introduction to Copulas , 2016 .

[14]  Jie Zhang,et al.  A Copula-Based Conditional Probabilistic Forecast Model for Wind Power Ramps , 2019, IEEE Transactions on Smart Grid.

[15]  Chunchi Wu,et al.  Time-Series Properties of Financial Series and Implications for Modeling , 1996 .

[16]  Claudia Czado,et al.  Analyzing Dependent Data with Vine Copulas , 2019, Lecture Notes in Statistics.

[17]  Goran Strbac,et al.  Using Vine Copulas to Generate Representative System States for Machine Learning , 2019, IEEE Transactions on Power Systems.

[18]  C. Genest,et al.  A Primer on Copulas for Count Data , 2007, ASTIN Bulletin.

[19]  C. Czado,et al.  Truncated regular vines in high dimensions with application to financial data , 2012 .

[20]  Andrew J. Patton.  Modelling Asymmetric Exchange Rate Dependence , 2006 .

[21]  H. Joe.  Families of $m$-variate distributions with given margins and $m(m-1)/2$ bivariate dependence parameters , 1996 .

[22]  Naseem Al Rahahleh,et al.  New evidence on fund performance in extreme events , 2019, International Journal of Managerial Finance.

[23]  Marno Verbeek,et al.  Selecting Copulas for Risk Management , 2007 .

[24]  Claudio Romano,et al.  Measuring and Optimizing Portfolio Credit Risk: A Copula-Based Approach , 2004 .

[25]  Gal Elidan,et al.  Copulas in Machine Learning , 2013 .

[26]  Olivier P. Faugeras,et al.  Inference for copula modeling of discrete data: a cautionary tale and some facts , 2017 .

[27]  Thong Nguyen-Huy,et al.  Copula-statistical precipitation forecasting model in Australia’s agro-ecological zones , 2017 .

[28]  Krzysztof Domino.  Multivariate cumulants in outlier detection for financial data analysis , 2018 .

[29]  Cuong Nguyen,et al.  Gold price and stock markets nexus under mixed-copulas , 2016 .

[30]  Harry Joe,et al.  Prediction based on conditional distributions of vine copulas , 2018, Comput. Stat. Data Anal..

[31]  M. Ishaq Bhatti,et al.  Co-movement measure of information transmission on international equity markets , 2017 .

[32]  M. Bhatti,et al.  Recent development in copula and its applications to the energy, forestry and environmental sciences , 2019, International Journal of Hydrogen Energy.

[33]  M. Sklar.  Fonctions de repartition a n dimensions et leurs marges , 1959 .

[34]  D. Dijk,et al.  Forecasting Volatility with Copula-Based Time Series Models , 2011 .

[35]  Andrew J. Patton.  Copula Methods for Forecasting Multivariate Time Series , 2013 .

[36]  Johan Segers,et al.  Extreme-value copulas , 2009, 0911.1015.

[37]  Jie Xu,et al.  Robust Estimation of Value-at-Risk through Distribution-Free and Parametric Approaches Using the Joint Severity and Frequency Model: Applications in Financial, Actuarial, and Natural Calamities Domains , 2017 .

[38]  Tiziano Bellini,et al.  Detecting atypical observations in financial data: the forward search for elliptical copulas , 2010, Adv. Data Anal. Classif..

[39]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[40]  Ruey S. Tsay,et al.  A Copula-based Approach to Option Pricing and Risk Assessment , 2008 .

[41]  Vijay P. Singh,et al.  A copula‐based precipitation forecasting model: Investigating the interdecadal modulation of ENSO's impacts on monthly precipitation , 2014 .