Our real-time forecast of the price of oil is based on the global oil market model proposed in Baumeister, Korobilis, and Lee (2022), "Energy Markets and Global Economic Conditions," published in the leading general-interest journal Review of Economics and Statistics.

The econometric framework is a Bayesian Vector AutoRegressive model (BVAR) that allows for time-varying volatility which is a key feature in energy markets.

Global consumption of crude oil and liquid fuels, global economic conditions, and OECD petroleum inventories.

The econometric model uses monthly data as they become available in real time. Data that arrive with a delay are nowcast at the end of each month when the forecast is generated. Taking advantage of backcasting techniques, our data span a long period of time starting in January 1973.