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.
5,922 documents
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…
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…
This work tests whether deep reinforcement learning can recover optimal trading strategies in environments that are simple to describe but mathematically challenging. The authors focus on deep deterministic policy gradient, a method that learns a policy for…
The note extends universal portfolio theory beyond constant-rebalanced strategies. It considers bilinear strategies, whose final growth depends linearly on each of two successive market return vectors, and compares them with the best such strategy selected…
The paper frames an investor’s problem as maximizing expected utility of wealth at a future horizon. It combines two kinds of positions: derivatives chosen once at the start and stock holdings adjusted dynamically over time. The setup allows the derivative…
The paper studies how to maximize long-run wealth growth when the market model is uncertain. It considers models constrained by a known asset region and instantaneous covariance, with an additional stability condition: the long-run occupancy measures…
AdaptiveTrend is a systematic cryptocurrency strategy that combines trend signals on six-hour intervals with monthly portfolio updates. It uses volatility-adjusted trailing stops, selects assets using rolling Sharpe ratios with market-capitalization filters,…
This survey reviews research on cryptocurrency trading, an area shaped by assets that share some features with traditional investments while retaining distinct market behavior. It covers trading systems and platforms, trading signals and strategies,…
This paper studies leakage in trading strategies generated from rank-dependent portfolio functions. Leakage refers to wealth lost when the stocks included in a portfolio are renewed or replaced. The authors derive a theoretical expression for this effect as…
This empirical study evaluates whether online portfolio strategies designed to exploit mean reversion perform well beyond benchmark datasets. It examines three approaches: passive aggressive mean reversion, online moving average reversion, and transaction…
This document presents a simulation-based method for managing bond portfolios over multiple periods while accounting for interest-rate risk and proportional trading costs. It models yield-curve movements with Dynamic Nelson-Siegel factors whose dynamics…
DeepUnifiedMom is a portfolio construction framework that combines time-series momentum signals across multiple time horizons. It uses multi-task learning with a multi-gate mixture-of-experts architecture to form unified portfolios, aiming to capture…
This paper proposes a theoretical control framework for market making in zero-maker-fee perpetual futures markets. A market maker chooses bid and ask spreads and manages inventory through hedging across two exchanges. The framework decomposes profit and loss…
This paper presents an optimization method for constructing a mean-reverting portfolio from a larger set of assets. It seeks portfolios whose spread is well described by an Ornstein–Uhlenbeck process, with process parameters estimated by maximum likelihood.…
This paper studies an investment and consumption problem in a pairs trading market where the spread between risky assets follows an Ornstein–Uhlenbeck process. It focuses on optimal strategies under power utility, bringing together portfolio decisions and…
The study examines the tails of Bitcoin return distributions across different time intervals and digital exchanges. It tests for common scaling behavior and estimates the power-law exponent, with attention to whether the tails imply a finite second moment.…
The document introduces virtual volatility as a measure of uncertainty in forecasts of the drift component of a random walk. It also presents the measure as a way to examine the stochastic behavior of a portfolio, with a reported example in which it helped…
This paper addresses portfolio management when market conditions change and the useful historical lookback window is not known in advance. It proposes a two-level process: construct portfolios using several candidate window sizes, then combine those…
This paper compares fixed stock-and-bond and stock, bond, and gold allocations with long-only dynamic portfolios that may also hold cash. The portfolios use public data, rebalance monthly, and are assessed using return, volatility, excess Sharpe ratio,…
This study combines language-model sentiment from financial news with macroeconomic indicators and technical signals to construct portfolios of Russell 2000 stocks. Its uncertainty-aware approach separates predicted risk into aleatoric and epistemic…
The study uses the timing proximity of trades across US stocks to define co-trading relationships and build dynamic networks. It applies spectral clustering to these networks to identify groups of stocks whose dependencies evolve over time, complementing…
The report studies long-horizon allocation between stocks and bonds when equity returns follow a mean-reverting process. It asks how this feature changes the risk-return trade-off and whether some strategies can make the portfolio’s terminal value behave as…
The document presents an approximation method for finding optimal strategies when trades incur proportional transaction costs. It represents the residual value function with a polynomial approximation, providing a way to handle problems where the exact…