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.
1,033 documents
The document asks why asset-price behavior appears to have changed little over roughly two centuries. It points to historical evidence of episodes with large price fluctuations and market inefficiency, then briefly reviews theoretical work and neurological…
This document presents a stochastic model of how opinions spread through a population containing trend-followers, opposers, and indifferent people. Reinforcement drives agents’ choices, while an unobserved trend process shapes those decisions. The framework…
This paper examines whether changes in the investor base of cryptocurrencies are associated with changes in their idiosyncratic volatility. It uses changes in subreddit follower counts as a proxy for shifts in investor base, then relates that measure to…
This study tests deep reinforcement learning as an execution overlay for pair trading in volatile cryptocurrency markets. Its framework first filters and ranks candidate pairs, then uses a Proximal Policy Optimization agent with a Long Short-Term Memory…
This work presents a generative pre-trained transformer for modeling financial time series as sequences of order messages. The model serves as an order-generation engine inside a discrete-event simulator, with the aim of reproducing limit order book dynamics…
The study examines high-frequency strategies for SPY options using five-minute observations collected over a one-month period. It calculates option Greeks and implied volatility, prices American options with a binomial tree, and estimates implied volatility…
The document describes an agent-based model of tick-by-tick price formation. Each agent’s buy and sell orders follow mutually exciting Hawkes processes, while mean-field interaction links activity across agents and creates correlated order volumes. This…
This work extends a latent order book model of market impact by allowing agents to exhibit mean-reverting behavior, a pattern commonly observed in markets. It studies the resulting order book dynamics using a mean-field assumption, which represents the book…
EverMine is a framework for testing whether research agents improve as they accumulate reusable skills, tools, and rules during long-horizon alpha discovery. It represents research state through the research history, current factor portfolio, and accumulated…
This paper compares two linear ways to describe how executed orders affect prices. The Transient Impact Model weights past market-order signs through a propagator, while the History Dependent Impact Model links price changes to the gap between an observed…
This paper studies information gaps in the Japanese stock market and asks whether investors outside conventional statistical arbitrage research can find useful signals in public online material. Its approach analyzes a large volume of blog data, then builds…
This document reports two studies of crypto rewards offered as conditional gifts to charitable donors. The first examines the Ukrainian government’s fundraising appeal accepting Bitcoin and Ethereum. When crypto rewards were present, the hourly donation…
This empirical study examines market impact in Bitcoin/USD using a dataset that reconstructs more than one million metaorders. It tests the square-root relationship between impact and order size, finding that the pattern holds across four orders of magnitude…
This document studies how temporary transaction costs affect an informed trader who knows an asset’s future value and a market maker who sets prices. It analyzes both a single-auction model and a continuous-time version, with the trader choosing trades to…
The paper combines statistically constructed factor models with stochastic control to study high-dimensional statistical arbitrage. Its model assumes mean-reverting residuals and shows how to build analytically market-neutral portfolios. It then derives…
The study proposes a reinforcement learning portfolio manager for a high-risk setting. It combines two-sided transactions with lending and borrowing, uses a profit-and-loss reward, and aims to improve downside risk management and capital use. Its…
The document develops a method for mean-variance hedging in general semimartingale markets. It frames hedging as a stochastic control problem and establishes that the control problem’s value process has a quadratic form. Three coefficient processes describe…
This study models the US Treasury yield curve as a latent continuous function observed only at a sparse set of maturities. Treating successive yield curves as a sparsely observed functional time series lets the authors examine how earlier macroeconomic…
The paper proposes a systematic investment approach intended to build wealth over a longer horizon through a sequence of short-term ETF purchase decisions. It combines evolutionary algorithms with a deep learning model to improve the timing of contributions…
StockGPT treats daily stock returns as sequences of numerical tokens and trains an autoregressive model to predict future returns. Its attention mechanism is intended to learn temporal patterns directly from return histories, avoiding the need to specify…
The document describes an empirical method for forecasting option prices by solving the Black–Scholes equation forward in time. Because the problem is ill-posed, the method uses regularization; the work formulates theorems concerning uniqueness, stability,…
The document examines option pricing and hedging on energy forwards when spot prices follow a geometric multifactor model with several mean-reversion rates. It focuses on how slow mean-reverting components affect valuation and hedge quality. The authors…
This document develops a framework for designing and comparing trend-following systems, including European, American, and time-series momentum approaches. For European systems, it derives a relationship between P&L, autocorrelation, and drift in…
This paper models how sequences of order-flow events affect asset prices. It extends earlier impact models that treated order flow as exogenous and summarized it mainly through pairwise correlations. Instead, it represents order flow directly as a discrete…