Research

This paper proposes a method for enhancing text-based forecasting models, with a specific focus on predicting crude oil prices. Utilizing advanced techniques, including pattern validation and attention mechanisms, the study demonstrates notable improvements in predictive power over traditional approaches. One key finding is that considering the full text of news articles, rather than limiting the analysis to news headlines, leads to significant gains in forecasting accuracy. Furthermore, the model featuring verb-noun and noun-verb collocation pattern validation consistently outperforms benchmarks and models based solely on news headlines across various forecasting horizons. The results suggest that the presence of such collocations as “price fell,” “prices tumbled,” and “price dropped” in crude-oil-related news articles is associated with a decrease in oil price returns. Additionally, integrating macroeconomic data with text-based features enhances predictive performance, demonstrating that combining structured economic indicators with textual features improves forecasting accuracy.

We use newspaper coverage to identify the supply and demand content of OPEC announcements. From 7,941 articles around 206 OPEC and OPEC+ announcements from 1971 to 2025, we construct separate text-based measures of supply and demand news without using futures prices. The supply measure produces macroeconomic responses similar to futures-based identification, while demand news generates distinct effects: both raise oil prices, but supply news contracts industrial production whereas demand news initially expands it.

Oil-market news can signal very different risks for banks exposed to oil-producing regions. This paper studies how newspaper-based oil narratives are associated with U.S. bank balance sheets. Supply-disruption narratives predict higher cumulative loan-loss provisions and net charge-offs, lower loan growth, lower deposit growth, and weaker short-run profitability among more oil-exposed banks. Demand-salience narratives move in the opposite direction, predicting stronger lending, deposit growth, and profitability, while provisions and charge-offs decline. These findings show that oil narratives do not have one banking effect: some oil stories become bank risk, while others coincide with expansion in oil-exposed banking markets

This paper examines oil-news attention as a predictor of crude oil price volatility. Using oil-related news articles from four major U.S. newspapers between 1988 and 2024, we construct monthly measures of oil-news attention and evaluate their out-of-sample forecasting power for WTI and Brent realized volatility. Adding oil-news attention to autoregressive (AR) and heterogeneous autoregressive (HAR) benchmarks significantly improves forecast accuracy, particularly for downside volatility. The improvements persist even when the benchmark already includes implied volatility from the VIX or the OVX, and they extend to 5% Value-at-Risk (VaR) forecasts. Consistent with the mixture-of-distributions hypothesis, what matters for forecasting oil-price risk is how much the press writes about oil, not what it says.

Asset bubbles can create the uncertainty that sustains them. We build an economy in which a gap between an asset's market price and the cost of covering its dividends finances project entry, expansion, and risk-taking. Those projects generate future uncertainty, which strains investors' capacity to bear risk and widens the bubble. This feedback can sustain several levels of bubbles and uncertainty even when shocks do not multiply on their own. Time structure determines the outcome. With a fixed ending date, equilibrium is unique regardless of feedback strength. When projects and uncertainty repeatedly renew, the same mechanism can produce multiple equilibria. Uncertainty shocks do not merely enlarge an existing bubble: by pushing risk beyond investors' capacity to bear it, they can cause one to form in the first place. The model identifies stable and unstable outcomes and remains valid in a fully closed economy.

This paper studies whether oil-market narratives have macroeconomic effects, and whether those effects differ depending on whether narratives convey directional information or uncertainty. I construct new monthly oil narrative sentiment and uncertainty indices from Wall Street Journal and New York Times articles from 1953 to 2022 and classify oil-related coverage into five narrative types. Using the type-specific narrative decomposition as external instruments in a Proxy-SVAR, I identify two structurally distinct narrative shocks. A bullish narrative sentiment shock raises the real oil price persistently, increases world industrial production, raises world oil production, and gradually passes through to consumer prices, consistent with a demand-side information channel. By contrast, an elevated narrative uncertainty shock reduces world oil production and world industrial production, while producing only a muted response of the real oil price and consumer prices, consistent with a supply-side real-options channel. When both shocks hit simultaneously, the combined effect produces a stagflationary pattern in which oil prices and inflation rise while production and activity respond weakly. These findings show that oil-market narratives contain macroeconomically relevant variation beyond physical oil shocks and that separating directional tone from ambiguity is essential for understanding how oil news transmits to the broader economy.

This research examines the welfare consequences of asset price bubble cycles, analyzing how boom–bust patterns in asset markets affect economic efficiency and social welfare.

This paper develops a novel forecasting methodology combining principal component analysis with weighted elastic net regularization to improve crude oil price return predictions using high-dimensional predictor sets.