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
This chapter surveys how wealth and trading behavior evolve in cryptocurrency systems that issue new coins through Proof of Stake. It first considers wealth accumulation without trading, with attention to how the protocol may affect decentralization. It then…
The paper extends eigenrisk parity’s isotropy principle into a portfolio allocation framework that combines mean-variance optimization with a tunable isotropy constraint. Isotropy spreads risk across eigenmodes and acts as a geometric regularizer intended to…
The paper presents a deep reinforcement learning agent for portfolio management that balances return seeking with risk restraint. Its target policy adjusts how strongly the agent favors the optimal action, using a tunable greediness parameter to encourage…
The document presents a framework for statistical arbitrage that combines asset pricing, time-series modeling, and portfolio optimization. It first uses conditional latent pricing factors to identify similar assets and construct residual portfolios, then…
This study examines how a household should invest while addressing a precautionary need with a random arrival time and an accumulation target due on a fixed date. The precautionary claim must be paid in full when it arrives if the household can afford it;…
The paper uses evolutionary optimization to compute rule-based trading strategies from financial sentiment data. The sentiment signal represents the online trading community’s bullish or bearish view of individual stocks and is sourced from StockTwits. The…
The paper introduces Volatility & Model Adaption Trade-off (VMAT), a strategy for multivariate pair trading. It addresses a gap the authors identify in prior work: research often refines trading rules, while portfolio management across multiple time series…
This paper makes the information used in mean-variance portfolio choice part of the optimization problem. It argues that standard portfolio analysis can be distorted by including variables unavailable at decision time or by mistaking shared variation for…
This paper studies two investors who choose mean-variance portfolios while competing over relative wealth. One investor knows the true stock dynamics; the other must infer them from observed market evolution. Their decisions are linked because each investor…
DeePM is presented as an end-to-end deep-learning system for managing diversified macro portfolios. Its design addresses asynchronous market data with a delay-based causal mechanism, constrains cross-asset relationships using an economics-informed graph, and…
The document outlines an intraday strategy for Chinese stocks under the market’s T+1 trading constraint. It combines forecasts from a multilayer perceptron (MLP) with Markowitz portfolio optimization, using published stock data from the Shanghai and Shenzhen…
The document studies how an investor with constant relative risk aversion should trade in a Black–Scholes market when they also hold a random endowment. It uses a duality approach to derive an explicit optimal trading strategy. The strategy is presented as…
This study tests whether sentence-level context improves news topic models used to build systematic asset-pricing factors. It compares Latent Dirichlet Allocation with a frozen sentence transformer followed by k-means clustering, using the same collection of…
This paper proposes a two-factor asset-pricing model for economies where firm capitalizations have a sufficiently heavy-tailed distribution. Alongside conventional market risk, the model includes diversification risk, proxied by an equally weighted…
The document presents a decade of evidence on applying trend-following strategies to cryptocurrency markets, beginning with Bitcoin’s early market history. It argues that crypto markets share characteristics with historical commodity markets and reports that…
This study presents SimStock, a framework for identifying similar stocks from time-series behavior rather than relying only on conventional sector or regional labels. It combines self-supervised learning with temporal domain generalization to form…
This note introduces Eigenrisk Parity, a portfolio construction approach that allocates risk evenly across the principal components of a covariance matrix. The broader idea is to use symmetry rather than rely on precise forecasts of returns and correlations.…
The paper offers a factor-based explanation for the low-volatility anomaly. It applies an Adaptive Multi-Factor model, estimated with the Groupwise Interpretable Basis Selection algorithm, to identify basis assets associated with low- and high-volatility…
This work compares financial mathematics and machine learning for dynamic portfolio optimization of two intertwined assets over a continuous-time, finite horizon. Its financial mathematics approach uses a cointelation model intended to combine short-term…
This note addresses the meaning of a self-financing trading strategy in quantitative finance. It places the concept at the intersection of stochastic calculus and financial trading, where misunderstandings may arise. The document’s stated aim is to remind…
This study builds a time-varying network of cryptocurrencies using two types of information: cross-predictability in returns and technological similarity. It applies a dynamic covariate-assisted spectral clustering method to estimate groups, or communities,…
The paper studies portfolio choice when the drift of a risky asset is not directly observed and depends on two hidden stochastic factors evolving at different speeds. It derives a filtered estimate of the latent mean-reversion level from the difference…
The paper derives an optimal portfolio strategy for an investor maximizing expected logarithmic utility when multivariate stock-return drift is hidden and evolves as an Ornstein–Uhlenbeck process. The investor learns about drift from observed returns and…
The paper applies machine-learning methods to forecast closing prices for a cryptocurrency index and nine constituent cryptocurrencies. It says that several algorithms and models are compared to identify the best-performing approach, and that the work is…