Working Paper: NBER ID: w29379
Authors: Charles W. Calomiris; Nida Akr Melek; Harry Mamaysky
Abstract: We study the performance of many traditional and novel, text-based variables for in-sample and out-of-sample forecasting of oil spot, futures, and energy company stock returns, and changes in oil volatility, production, and inventories. After controlling for small-sample biases, we find evidence of in-sample predictability. Our text measures, derived using energy news articles, hold their own against traditional variables. While we cannot identify ex-ante rules for selecting successful out-of-sample forecasters, an analysis of all possible two-variable models reveals out-of-sample performance above that expected under random variation. Our findings provide new directions for identifying robust forecasting models for oil markets, and beyond.
Keywords: Oil Market; Forecasting; Text Analysis; Predictability
JEL Codes: C52; G10; G12; G14; G17; Q47
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Traditional predictors, such as macroeconomic indicators and financial metrics (G17) | Oil spot and futures returns (G13) |
| NLP measures derived from energy news articles (C45) | Oil spot and futures returns (G13) |
| Traditional predictors and NLP measures (C52) | Oil market outcomes (L71) |
| Forward selection model (C52) | Parsimonious set of forecasting variables (C29) |
| In-sample predictability (C53) | Out-of-sample forecasting success (C53) |
| Model instability (C62) | Future forecasting success (G17) |