Clustering Commodity Markets in Space and Time: Clarifying Returns, Volatility, and Trading Regimes Through Unsupervised Machine Learning

90 Pages Posted: 31 Mar 2021

See all articles by James Ming Chen

James Ming Chen

Michigan State University - College of Law

Mobeen Ur Rehman

Institute of Business Research, University of Economics; South Ural State University, Russian Federation

Xuan Vinh Vo

University of Economics Ho Chi Minh City; French-Vietnamese Center for Management Education (CFVG) in Ho Chi Minh City

Date Written: February 23, 2021

Abstract

Unsupervised machine learning can interpret logarithmic returns and conditional volatility in commodity markets. k-means and hierarchical clustering can generate a financial ontology of markets for fuels, precious and base metals, and agricultural commodities. Manifold learning methods such as multidimensional scaling (MDS) and t-distributed stochastic neighbor embedding (t-SNE) enable the visualization of comovement and other financial relationships in three dimensions.

Different methods of unsupervised learning excel at different tasks. k-means clustering based on logarithmic returns works well with MDS to classify commodities and to create a spatial ontology of commodities trading, A strikingly different application involves k-means clustering of the matrix transpose, such that conditional volatility is evaluated by trading date rather than by commodity. This approach can isolate the two most calamitous temporal regimes of the past two decades: the global financial crisis of 2008-09 and the immediate reaction to the Covid-19 pandemic. Temporal clustering of trading days, unlike the corresponding spatial task of clustering commodities, is better visualized through t-SNE than through MDS.

Keywords: Commodities, commodity markets, precious metals, base metals, energy markets, agricultural markets, machine learning, unsupervised learning, GARCH, clustering, k-means clustering, hierarchical agglomerative clustering, multidimensional scaling, MDS, t-distributed stochastic neighbor embedding, t-SNE

JEL Classification: C38, C65, Q02

Suggested Citation

Chen, James Ming and Rehman, Mobeen Ur and Vo, Xuan Vinh, Clustering Commodity Markets in Space and Time: Clarifying Returns, Volatility, and Trading Regimes Through Unsupervised Machine Learning (February 23, 2021). Available at SSRN: https://ssrn.com/abstract=3791138 or http://dx.doi.org/10.2139/ssrn.3791138

James Ming Chen (Contact Author)

Michigan State University - College of Law ( email )

318 Law College Building
East Lansing, MI 48824-1300
United States

Mobeen Ur Rehman

Institute of Business Research, University of Economics ( email )

Ho Chi Minh City
Vietnam

South Ural State University, Russian Federation ( email )

76, Lenin Prospekt
Chelyabinsk
Russia

Xuan Vinh Vo

University of Economics Ho Chi Minh City ( email )

Ho Chi Minh City, Ho Chi Minh City
Vietnam

HOME PAGE: http://www.ueh.edu.vn

French-Vietnamese Center for Management Education (CFVG) in Ho Chi Minh City

91 Ba Thang Hai Street
District 10
Ho Chi Minh City
Vietnam

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
85
Abstract Views
511
rank
348,225
PlumX Metrics