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.
4,613 documents
This paper formulates an insurer’s joint reinsurance and dividend decisions as a stochastic control problem. The risk process is Brownian, with drift and volatility changing through endogenous regime switching. Dividend payments incur fixed transaction…
This study formulates portfolio trading as a Markov decision process and trains an agent with deep Q-learning to choose allocations across multiple assets. Its action space is discrete and combinatorial: for each asset, the agent selects a trading direction…
The document presents AutoQuant, a rule-based decision-support framework for selecting trading configurations in cryptocurrency perpetual futures. It makes execution timing, funding treatment, fees, slippage, and feasibility requirements explicit, then uses…
This paper designs a spouse-protected tontine for household retirement income. A first death changes the household’s status without triggering a transfer from the pool; the account stays with the contract and can fund a spouse-only continuation if the…
The document derives how the continuous-time Kelly fraction changes when an asset’s long-run price follows a power law and its return variance declines over time. With zero risk-free return, the allocation depends on the growth exponent, the volatility…
The document presents a federated learning approach for estimating borrower income when lenders cannot combine raw applicant records. Institutions train a shared model while keeping data local. Its evaluation simulates a consortium of state-based clients…
This study tests a gold futures strategy built from smoothed trend and momentum state variables. A rolling walk-forward process trains on a decade of data and tests on successive six-month periods across 2015–2025. The signal is translated into positions…
The document describes a computational approach for a trader seeking portfolio wealth outcomes that satisfy a convex risk measure. The proposed black-box method takes the joint distribution of stock prices and the chosen risk measure as inputs, then…
This document studies how to manage positions across several mean-reverting assets, including whether an asset with no mean reversion can still belong in the portfolio. It formulates position management as an optimal control problem for an investor with…
This paper studies Bertram’s optimal strategy for a pair of cointegrated assets whose price difference follows an Ornstein–Uhlenbeck process. The baseline objective is to maximize expected profit per unit time. The authors generalize the setup by…
This study applies deep reinforcement learning to continuous futures trading, examining both discrete and continuous action spaces. It incorporates volatility scaling into reward functions so that position sizes respond to market volatility. The algorithms…
This study presents a historical-data-based deep learning framework for systematic options trading that incorporates portfolio risk sensitivities directly into training. Its objective combines a performance-oriented loss with a differentiable penalty for…
The document explains time series momentum as a strategy that uses each instrument’s own past return, rather than ranking assets against one another. Its central signal is the sign of the prior 12-month excess return: go long when positive and short when…
The document explains how investors with limited capital can implement equity momentum without holding hundreds of stocks. Its example ranks UK-listed companies by their returns over the prior 12 months, excludes the smallest quarter of firms for liquidity,…
This trend-following approach compares three moving averages, typically ordered from a shorter to a longer lookback. A selected pair crossing can trigger a market entry; the description allows trading the first average against either or both longer averages.…
This strategy combines crossovers between Hann-filtered price series with market-structure signals. It tracks swing highs and lows, flags breaks of those levels, and uses a structural trend state to avoid entries against a confirmed direction. An ADX range…
This opening installment considers how a trader with limited capital might allocate across futures when contracts cannot be traded fractionally. The author frames the challenge as a portfolio optimization problem: a small account cannot spread capital over…
The document describes an automated RSI strategy that uses H1 signals with D1 confirmation. A buy requires RSI on both timeframes to be above the oversold threshold and rising relative to its previous reading; a sell requires RSI below the overbought…
This document explains an expert advisor approach for moving stop losses as a trade becomes profitable. The user sets whether trailing is enabled, a profit distance at which trailing begins, a step size, and a magic number. On each tick, the routine scans…
This article presents a reusable function library for MQL4 Expert Advisors. Its order-opening functions support stop-loss and take-profit levels specified either as distances from entry or as absolute chart prices. Separate function variants address brokers…
This guide presents ten broad principles for participating in cryptocurrency markets. It emphasizes limiting exposure to money an investor can afford to lose, building positions through scheduled dollar-cost averaging, researching token utility and project…
The document explains how leverage in gold contracts for difference changes the margin needed to hold a position. It compares the approximate margin requirements at 100x, 500x, and 800x leverage, illustrating how higher leverage can leave more capital…
The guide explains how holders of eligible tokenized U.S. stocks and ETFs can pledge them as collateral for crypto loans rather than selling their stock-linked exposure. It outlines the borrowing sequence, possible uses of borrowed assets, and the role of…
A long strangle buys an out-of-the-money call and put on the same stock. The call can gain if the stock rises substantially, while the put can gain if it falls; the buyer’s maximum loss is the premiums paid. The document proposes opening these positions in…