Library

A small set of books and papers I keep returning to. Updated when something earns a place.

• The Misbehavior of Markets, Benoit Mandelbrot & Richard Hudson, 2004 Mandelbrot applies fractal geometry to price series and argues that markets are far more volatile than Gaussian models permit; essential reading before building any risk model that treats variance as an adequate summary of uncertainty.
• Reminiscences of a Stock Operator, Edwin Lefèvre, 1923 A fictionalized account of Jesse Livermore's career that most people in markets read eventually; what endures past the era-specific trades is the extended account of patience, of waiting until conditions are right rather than manufacturing reasons to act.
• When Genius Failed, Roger Lowenstein, 2000 The LTCM collapse reconstructed at the level of positions and people; more instructive than the general lesson about leverage is the specific account of how arbitrage spreads widened exactly when everyone holding them needed them to converge.
• The Endurance, Alfred Lansing, 1959 Lansing reconstructed Shackleton's 1914 Antarctic expedition from diaries and interviews; what stays with you is how a leader keeps people functional when all the original goals are gone and survival is the only one left.
• The Myth of Sisyphus, Albert Camus, 1942 Camus takes seriously the possibility that the universe offers no justification for human effort and concludes not with despair but with a decision to revolt rather than submit; one of the few works of philosophy that actually changes the texture of daily experience.
• Ficciones, Jorge Luis Borges, 1944 Borges builds stories out of logical paradoxes and combinatorial infinities rather than character or event; The Library of Babel reads differently once you have spent time thinking seriously about information, entropy, and what it means for something to be possible.
• Gödel, Escher, Bach, Douglas Hofstadter, 1979 An extended argument that self-reference is the mechanism behind consciousness and meaning, threaded through Bach's counterpoint, Escher's impossible figures, and the structure of Gödel's incompleteness proof; the kind of book that changes what you notice in unrelated fields.
• The Feynman Lectures on Physics, Feynman, Leighton & Sands, 1964 Three volumes of undergraduate physics developed for Caltech's strongest incoming students; the lasting value is not the material, which you can find anywhere, but Feynman's refusal to stop until he can give a physical account of what is actually happening rather than just what the formalism says.
• A Mathematician's Apology, G. H. Hardy, 1940 Hardy defends the pursuit of pure mathematics as an aesthetic calling with no obligation to justify itself through usefulness; written when he knew his creative years were behind him, and that self-awareness gives it a weight most books about mathematics do not have.
• Critique of Pure Reason, Immanuel Kant, 1781 Kant's argument that the mind imposes structure on experience rather than passively receiving it changed how philosophy frames every subsequent question about knowledge; slow going, but worth the effort on the Transcendental Aesthetic alone.

• Time Series Momentum, Moskowitz, Ooi & Pedersen, 2012 · Journal of Financial Economics Documents that past returns predict future returns for each asset in isolation across 58 futures markets; the cleanest empirical foundation for trend-following strategies and a useful starting point for thinking about when and why momentum works.
• A Century of Evidence on Trend-Following Investing, Hurst, Ooi & Pedersen, 2017 · Journal of Portfolio Management Extends time-series momentum evidence back to 1880 across equities, bonds, currencies, and commodities; the long out-of-sample history is the main response to anyone who worries that the Moskowitz-Ooi-Pedersen result is an artifact of a short sample period.
• Continuous Auctions and Insider Trading, A. S. Kyle, 1985 · Econometrica Introduces the model where an informed trader optimally disguises order flow in the noise from liquidity traders, giving us the lambda that measures how much prices move per unit of order flow and grounding most of what came after in market microstructure.
• Optimal Execution of Portfolio Transactions, Almgren & Chriss, 2001 · Journal of Risk Derives the efficient frontier between expected transaction cost and cost variance when liquidating a large position, giving a rigorous framework for the tradeoff between trading slowly to minimize market impact and trading quickly to minimize timing risk.
• Bid, Ask and Transaction Prices in a Specialist Market, Glosten & Milgrom, 1985 · Journal of Financial Economics Shows that the bid-ask spread exists partly as compensation for the risk of trading against someone who knows more than you; the adverse selection framing here is the lens through which most practitioners still think about quoting and inventory risk.
• The Components of the Bid-Ask Spread, Huang & Stoll, 1997 · Review of Financial Studies Decomposes the spread into order processing costs, inventory holding costs, and adverse selection; the decomposition is what makes the spread useful as a diagnostic rather than just a transaction cost to minimize.
• A Simple Implicit Measure of the Effective Bid-Ask Spread, Richard Roll, 1984 · Journal of Finance Derives a spread estimator from the serial covariance of price changes alone, requiring no order book data; elegant in its parsimony and still the go-to when you only have trade prices to work with.
• The Market for Lemons, George Akerlof, 1970 · Quarterly Journal of Economics Shows that asymmetric information alone is enough to cause markets for high-quality goods to collapse entirely; the used-car example is the memorable framing but the structure of the argument applies anywhere one side of a transaction knows substantially more than the other.

• Situational Awareness, Leopold Aschenbrenner, 2024 The most seriously argued public document on the near-term trajectory of AI development written by someone with firsthand exposure to frontier training; worth reading regardless of whether you accept the conclusions, because the framing shapes how you read almost everything else on the subject.
• You Can't Predict. You Can Prepare., Howard Marks, 2001 · Oaktree Capital memo Marks argues that the right response to uncertainty is not better forecasting but building portfolios that can survive the scenarios you can imagine and some you cannot; the September 2001 timing gives the memo a weight that a normal market commentary would not have.
• Meditations on Moloch, Scott Alexander, 2014 · Slate Star Codex A long essay on coordination failure that works through the structure of multipolar traps, situations where individually rational behavior produces collective outcomes nobody chose; the Ginsberg framing is memorable but the actual argument about optimization and values is the reason to read it.