Oil price volatility predictability based on global economic conditions

This study aims to examine the forecasting ability of five global economic activity proxies for WTI crude oil price volatility and construct a new index to improve the accuracy of WTI crude oil price volatility forecasts. We focus on the Global Economic Conditions Index (GECON) derived from 16 indicators related to real economic activity and adopt the autoregressive (AR) framework, along with three common indexes constructed by three dimen-sionality reduction approaches (scaled principal component analysis (sPCA), principal component analysis (PCA) and partial least squares (PLS)). The out-of-sample results show that the model incorporating the Global Eco-nomic Conditions Index (AR-GECON) has the strongest predictive power among the five global economic proxy models. More importantly, our newly constructed PLS model outperforms all the other forecasting models, including AR-GECON.

Automated summary

This study aims to examine the forecasting ability of five global economic activity proxies for WTI crude oil price volatility and construct a new index to improve the accuracy of WTI crude oil price volatility forecasts. The results show that the Global Economic Conditions Index (GECON) and partial least squares (PLS) model have strong predictive power for WTI crude oil price volatility.