Research
Research notes and preprints, plus a reading list of 18 papers — the ones we think are worth the evening, from the equilibrium arguments the whole field rests on to what deep learning has actually established about price formation.
Papers
No papers match that search.
LLM-Extracted Earnings Sentiment as an Alpha Factor
We evaluate frontier LLMs as extractors of forward-looking sentiment from earnings call transcripts. Semantic delta across consecutive calls achieves 3-day alpha of 1.8% with Sharpe 1.72 OOS.
Yield Curve Inversion as a Regime Classifier for Equity Factor Rotation
We propose a hidden Markov model that conditions equity factor exposures on yield curve shape. Under inversion, momentum and quality factors dominate; under normalization, value and low-volatility outperform by 340bps annualized.
Recommended reading · 18 papers
Papers we come back to. Each note says why it earns the spot rather than restating the abstract — read the notes to decide what to open, then read the paper. Every link goes to the canonical source: a DOI, arXiv, SSRN, or the publisher's own PDF. Some sit behind a paywall; a preprint is usually a search away.
Trading Strategies and Market Microstructure: Evidence from a Prediction Market
David M. Rothschild, Rajiv Sethi
Account-level data on 6,300 traders in the Intrade 2012 election market — a whole market ecology laid bare, including evidence of manipulation by one large trader. Rare visibility into who is actually on the other side.
Foundations of Factor Investing
Jennifer Bender, Remy Briand, Dimitris Melas, Raman Aylur Subramanian
The clearest bridge from academic factor papers to what an index provider actually builds. Read it when you need to explain factors to someone who allocates capital.
The Value Premium and the CAPM
Eugene F. Fama, Kenneth R. French
The direct sequel to Sharpe: value beats the market, and the CAPM beta cannot explain it. Pairs with the 1993 three-factor paper below.
A Simple Approach to the Theory of Asset Prices
Riccardo Cesari, Carlo D'Adda
Prices assets as bundles of moment-characteristics without assuming expected utility. A useful antidote if you have only ever seen pricing derived one way.
Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk
William F. Sharpe
The CAPM. Read it for the argument, not the result: Sharpe derives a single price of risk from equilibrium alone, and every factor model since is a fight about what he left out.
Empirical Asset Pricing via Machine Learning
Shihao Gu, Bryan Kelly, Dacheng Xiu
The reference comparison of ML methods for return prediction. The interesting result is not that trees and nets win, but that their gains come from allowing interactions and nonlinearity — not from more data.
DeepLOB: Deep Convolutional Neural Networks for Limit Order Books
Zihao Zhang, Stefan Zohren, Stephen Roberts
A concrete, reproducible architecture for learning directly from raw order book state. Good starting point if you want to move past hand-engineered microstructure features.
Universal Features of Price Formation in Financial Markets: Perspectives from Deep Learning
Justin Sirignano, Rama Cont
A single deep network trained across stocks beats per-stock models, which is evidence that price formation has a universal component. The strongest argument for pooling data that we know of.
… and the Cross-Section of Expected Returns
Campbell R. Harvey, Yan Liu, Heqing Zhu
Hundreds of published factors, one multiple-testing correction, and most of them stop being significant. The single most useful paper for staying honest about your own research.
The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting and Non-Normality
David H. Bailey, Marcos López de Prado
How to discount a Sharpe ratio for the number of trials it took to find it. Practical, and it usually deflates your favourite strategy more than you expect.
Optimal Execution of Portfolio Transactions
Robert Almgren, Neil Chriss
The trade-off between market impact and timing risk, made tractable. If your backtest assumes fills at mid, this is the paper that tells you what you owe.
On Persistence in Mutual Fund Performance
Mark M. Carhart
Adds momentum to the three-factor model and, in doing so, explains away most of what looked like fund manager skill. A lesson in attribution before admiration.
Common Risk Factors in the Returns on Stocks and Bonds
Eugene F. Fama, Kenneth R. French
Size and value as priced factors. The three-factor model is the baseline any new signal has to beat before it is worth a second look.
Returns to Buying Winners and Selling Losers
Narasimhan Jegadeesh, Sheridan Titman
Momentum: the anomaly that refused to be explained away and still has not been. Note how carefully they handle overlapping holding periods — the method matters as much as the finding.
Continuous Auctions and Insider Trading
Albert S. Kyle
Where market impact comes from. Kyle’s lambda is still the unit in which execution desks think about how much your order moves the price.
The Pricing of Options and Corporate Liabilities
Fischer Black, Myron Scholes
Replication, not forecasting. The insight that a dynamic hedge pins down a price is the template for most modern derivatives work — and its assumptions are the standard list of what breaks in practice.
Efficient Capital Markets: A Review of Theory and Empirical Work
Eugene F. Fama
The joint-hypothesis problem lives here: you can never test market efficiency without also testing your pricing model. Worth rereading whenever a backtest looks too good.
Portfolio Selection
Harry Markowitz
Fourteen pages that turned investing into an optimization problem. Everything downstream — risk models, covariance shrinkage, risk parity — is an argument with this paper.