NEWS & EVENTS

By Shawn Hutchins

Amid today’s demanding market environment and economic uncertainty over inflation, statisticians are using complex quantitative frameworks to analyze the dynamics and establish data-backed insights and decisions. 

Energy sector commodities can be volatile. In recent years, crude oil prices have plummeted and then soared due to pandemic lockdowns, an epic oil glut and severe global supply disruptions. Such challenges have downstream effects and have had a lot to do with rising inflation. 

At a time when non-renewable resources are under increasing stress, Rice University statisticians are analyzing the inextricable linkage commonly known as the water-energy nexus.

“From fracking and the production of petroleum to nuclear power and the distilling of biofuels, almost every power source demands water,” said Katherine Ensor, Rice’s Noah G. Harding Professor of Statistics and director of the Center for Computational Finance and Economic Systems (CoFES). “Water and energy are highly interdependent and both sectors are set to intensify in the coming future due to rising demands and constraints.”

A new paper, published on August 8 in the Indian journal Sankhya by Ensor and Rice alumna Kim Raath, quantifies the dynamic financial interactions of the water-energy nexus. Within a series of well-known stochastic volatility models in time series data, the research augments forecasts by applying a unique WaveL2E thresholding method. 

Commodity marketplaces are complex and difficult to forecast due to a spectrum of global events and participants with an array of objectives. The WaveL2E method is a mathematical framework, developed as part of Raath’s doctoral research, that combines Morlet’s continuous wavelet transform to denoise, or smooth, random data fluctuations in non-stationary time series.

The work is the basis of ongoing CoFES research using a CoFESWave R package, which implements the thresholds introduced in the paper. 

“This wavet-based method allows us to visualize structure in multidimensional data over different scales. In commodity analysis, WaveL2E is responsive to volatility in a time-frequency analysis and allows us to deconstruct and rigorously characterize data structure in different investment horizons and extract optimal components,” said Raath, who completed a doctorate in statistics under Ensor’s direction and a professional master in economics from Rice in 2020. 

“These estimates allow an opportunity to simultaneously model the dynamic behavior of the signal in the time series and the volatility of the noise variance,” added Raath. 

Raath is the CTO of Imperative Global, a market-leading carbon offset project developer and operator. She was previously CEO and co-founder of Topl, a purpose-built blockchain infrastructure designed to increase coordination and efficiency in ethical and sustainable value chain ecosystems and unlock innovative ESG financing mechanisms across industries. She is also a board advisor to Xpansiv, the leading global market for ESG-inclusive commodities, a member of the CNBC ESG Council, and an advisory board member of Rice’s Center for Computational Finance and Economic Systems (CoFES).

Raath and Ensor tested the WaveL2E method by examining the dynamic economic behavior of the water-energy nexus from June 6, 2007 to January 26, 2018. The timeline included two notable periods when prices bubbled and collapsed: the 2008 global financial crisis and the subsequent 2014-2016 global oil glut. 

The research pointed out the complexity and interconnected relationship of the commodity markets that make up the water-energy nexus by analyzing 2,683 data points for each series of the Energy Select Sector SPDR Fund (XLE) and the Invesco S&P Global Water Index ETF (CGW). The SPDR S&P 500 ETF Trust, which is widely recognized and represents U.S. market behavior and trading volume, was used to show the comparative analysis between traditional stochastic volatility methods and the WaveL2E method.

After an overview of the accuracy of the proposed models, Raath and Ensor concluded that wavelets not only improve forecasting quality but are also promising tools for returns series that contain a lot of volatility clustering. 

“The technique also presents the importance of continued quantitative research in the water-energy financial relationship and in evaluating responsible investment approaches,” Ensor said.