Jesus Villota Miranda

Large traders usually split orders to reduce price impact and avoid attracting attention. The largest trades in equity options increasingly do the opposite. Trades above the 99th percentile of the trade-size distribution for the same underlying almost tripled after January 2020. These trades are visible, use much of the displayed depth, and move option prices by more than 2 percent at the trade. Their execution costs did not fall after 2020. What changed was the audience. Retail investors gained much easier access to the options tape, while option-flow services began identifying and interpreting large trades in real time. Our proxies for retail option trading in the same underlying rise for hours after a whale trade, and the response is about four times larger after 2020. The additional trading arrives disproportionately on the whale's side of the market. Less of the initial price impact reverses, and it reverses least for the prints that draw the most retail trading. Large visible trades are therefore followed by both more retail trading and more persistent price effects.

The profitability of traditional pairs trading, a market-neutral strategy based on identifying and exploiting temporary price divergences between two historically related stocks, has significantly eroded over the past two decades. This paper argues that the decline is not due to a failure of the underlying principle of relative-value arbitrage, but rather to the restrictive 1-to-1 nature of the conventional methodology. We propose a generalization of pairs trading by replacing the single partner stock with a replicating portfolio constructed as a linear combination of multiple securities. Our approach utilizes LASSO to create a parsimonious and tradable portfolio that collectively mimics the price behavior of a target asset, thereby creating a more robust and flexible substitute. Our empirical analysis, conducted on U.S. equity data from 1962 to the present, demonstrates that while the profitability of the classic approach has decayed, our generalized strategy consistently delivers significant excess returns, particularly in the post-2000 period.

Markets do not always efficiently incorporate news, particularly when information is complex or ambiguous. Traditional text analysis methods fail to capture the economic structure of information and its firm-specific implications. This paper proposes a novel methodology that guides LLMs to systematically identify and classify firm-specific economic shocks in news articles according to their type, magnitude, and direction. This economically-informed classification allows for a more nuanced understanding of how markets process complex information. Using a simple trading strategy, we demonstrate that our LLM-based classification significantly outperforms a benchmark based on clustering vector embeddings, generating consistent profits out-of-sample while maintaining transparent and durable trading signals. The results suggest that LLMs, when properly guided by economic frameworks, can effectively identify persistent patterns in how markets react to different types of firm-specific news.

This paper focuses on the economics of vaccination using a game-theoretic model combined with an epidemiological SIR model that reproduces the infection dynamics of a generic disease. We characterize the equilibrium individual vaccination rate and show that it is below the rate compatible with herd immunity due to externalities that individuals do not internalize. We analyze three public policies: informational campaigns to reduce vaccination disutility, monetary payments to vaccinated individuals, and measures to increase the disutility of non-vaccination. We find that the optimal public policy should consist only of informational campaigns if they are sufficiently effective, or a combination of informational campaigns and monetary incentives otherwise. Surprisingly, vaccine passports or other restrictions on the non-vaccinated are not desirable.