Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
195 documents
The article distinguishes risk premia, which compensate traders for bearing unwanted risks, from inefficiencies caused by participants who must trade for reasons other than price. Forced liquidations, redemptions, reporting practices, index changes, and…
The article reviews a strong year for diversified systematic trading and focuses on a bond strategy that buys shortly before month-end, sells at month-end, and re-enters after a few days. The author describes institutional rebalancing and portfolio…
The document frames trading as judging whether an asset is mispriced, then competing with others who may recognize the same opportunity. Expected buying pressure, for example, can be reflected in the price before a trader is able to act. This competition…
The document explains how the Graphical Lasso estimates a sparse inverse covariance matrix from stock data. After scaling its off-diagonal entries, the method derives partial correlations, which describe the relationship between two stocks while accounting…
This tutorial explains how to calculate the expiration profit or loss of a long call or put. It distinguishes an option’s intrinsic value at expiration from the position’s net result by subtracting the premium paid. Worked examples show a call finishing…
The document summarizes proposed cross-sectional signals for judging whether equity options are relatively cheap or expensive. Its central comparison is implied volatility against volatility that later realizes: options may be candidates to buy when implied…
This guide outlines the capabilities and working practices needed to develop algorithmic trading systems. It highlights programming, statistics, and risk management, with Python and R presented as useful research tools. It also gives criteria for choosing a…
The document frames consistent participation in markets as a way to grow capital over time. It points to the time value of money and the no-arbitrage principle as the main ideas for understanding how investments can earn more than a baseline return, though…
This quiz presents a compact lesson about the tension between market efficiency and noisy price movements. Its central implication is that even sound trading decisions can feel messy, imprecise, or uncomfortable because a trader’s edge may be small relative…
This article argues that programming simulations can make statistical questions more intuitive than relying solely on classical formulas. It illustrates the approach with roulette: under a stated single-number win probability, repeated simulated sequences…
This article presents a practical workflow for exploratory research on SPY using QuantConnect. It examines daily return distributions, compares them with a normal distribution, looks for possible calendar and intraday seasonal patterns, and measures return…
The article examines whether US election dates coincide with unusual S&P 500 returns. It describes aligning historical index returns to the nearest election, grouping observations by days before or after election day, and comparing average returns across the…
This article focuses on selecting stock pairs for statistical arbitrage. It argues that finding pairs whose prices reliably diverge and reconverge matters more than the details of hedge-ratio estimation or other implementation models. Historical correlation…
This article outlines common ways systematic trading experiments can mislead. It names look-ahead bias, where a test uses information unavailable at the time of a trade; overfitting, where rules or parameters are tuned to historical noise; and data-mining or…
This introductory page presents a research philosophy for independent systematic traders: begin by identifying a plausible market edge, then use tools such as backtesting to investigate it. A backtest can show how a set of rules performed historically, but…
The article defines a trading edge as positive expected value: across many trades, the probability-weighted gains should exceed the losses. A strategy can lose often and still have an edge, or win frequently while carrying occasional losses large enough to…
This article weighs gold’s theoretical status against its observed portfolio behavior. Since gold produces no cash flow and has no clear cash-flow-based valuation anchor, the author argues it does not fit a conventional academic account of a risk premium. In…
This article argues that traders should not make statistical significance the sole test for acting on an idea. In markets with short histories, rare events, or structural changes, a useful edge may not have enough observations to produce a reliable p-value…
This tutorial describes how to connect the R statistical environment to Zorro, allowing a Zorro script to send market data to R, run R computations, and retrieve results. It outlines configuring the R installation, starting and checking an R session, and…
The article presents pairs trading as taking opposite positions in correlated assets when their relative prices diverge, with the expectation that the relationship will move back toward its mean. It questions the routine use of price regression to estimate a…
The article argues that AI makes it easy to generate and test trading rules, but that speed also encourages data mining. Repeatedly changing parameters, filters, timeframes, or asset universes amounts to many hypothesis tests; a strong historical result can…
The article addresses how to distinguish a durable strategy effect from luck, while acknowledging that certainty is impossible. It recommends starting with a credible economic explanation, such as compensation for bearing risk or a structural imbalance in…
This tutorial demonstrates a workflow for obtaining cryptocurrency listings, market capitalization, trading volume, and daily historical prices through the CryptoCompare API. It batches coin queries, ranks assets by reported market capitalization, removes…
The article argues that language models are unreliable for discovering trading edges because of three problems: conventional trading advice dominates their training data, models struggle to retrieve the latest value after repeated updates, and their…