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.
3,481 documents
This research summary compares common ways to divide macroeconomic cycles: the growth-and-inflation framework often called the Merrill Lynch clock, monetary-credit cycles, and monetary-economic cycles. It argues that these approaches revolve around money,…
The post describes a problem loading a saved StockRanker model into a live trading environment. The author reports that prediction fails with a JSON parsing error, even though the model works on the platform in the development environment. This is a report…
This discussion explains why assets under management and quantitative strategy returns may be related without having a simple linear relationship. It argues that capacity depends on the strategy: too little capital may prevent a portfolio from meeting its…
The document describes Chris Camillo’s social-arbitrage approach: watch social media and everyday consumer behavior for emerging trends that may not yet be reflected in market expectations, then connect those signals to publicly traded companies. It…
This research summary describes factors built from intraday snapshot data to measure the relationship between stock prices and trading volume. The reported premise is that price-volume divergence may precede higher returns whether prices are rising or…
This report describes a stock-selection approach that optimizes portfolio weights while constraining market-cap, industry, and style exposures. Its goal is to retain exposure to alpha factors while limiting unwanted variation from other risk factors. The…
The article challenges the idea that US regulators define high-frequency trading through a universal orders-per-second limit. It says figures repeated in public debate come from academic or market research definitions rather than an official regulatory…
This note outlines two ways to include industry or sector membership in stock factor research. One approach treats each industry or sector as a grouping variable and compares the cumulative returns of its constituent stocks, effectively studying group…
This 2020 research summary analyzes the Silver Hua CSI Innovative Drug Industry ETF and its underlying index. The index selects up to 50 listed companies associated with innovative drug research, ranked by market capitalization. The report discusses the…
The document introduces the case for identifying actively managed industry theme funds in China’s A-share market. It describes pronounced differences in industry returns and rotation over the prior decade, with some sectors producing strong gains even when…
This survey introduces supervised, unsupervised, semi-supervised, and reinforcement learning, then outlines common methods such as neural networks, support vector machines, nearest neighbors, clustering, and Bayesian networks. It places machine learning in…
This survey introduces supervised, unsupervised, semi-supervised, and reinforcement learning, then outlines common methods such as neural networks, support vector machines, nearest neighbors, clustering, and Bayesian networks. It places machine learning in…
This article explains a holdings-based panel regression approach to evaluating mutual fund performance. With stock fixed effects, a conventional performance measure can be separated into time-series ability (TSA), reflecting whether changing portfolio…
This review examines whether published indicators of future mutual-fund performance retain their predictive power outside the periods in which they were studied. It compares high-minus-low portfolio alpha spreads and cross-sectional regression slopes across…
This paper guide summarizes research on convolutional networks, model efficiency, and image segmentation. It describes channel attention through feature recalibration, depthwise separable convolutions for reducing computation, and encoder-decoder or…
The article explains idiosyncratic volatility for individual stocks and extends the idea to residual volatility for indices and commodity futures. It argues that elevated volatility is more useful for anticipating the size of future moves or divergence from…
This brief forum exchange concerns an SQL error encountered after adding a z-score standardization function to factor data. The reported issue is that a query joining tables cannot determine which table supplies a column named date. The response recommends…
This tutorial outlines a Java client that connects to a streaming market data service over WebSocket, authenticates with an API key, and subscribes to selected instruments and data types. It explains the role of the Java endpoint annotations and session, and…
This research roundup summarizes two studies. The first compares the Fama–French five-factor model with a version that adds momentum, then considers replacing the standard value factor with a more immediate value measure. The summary says the six-factor…
This article describes four alleged ways quantitative traders affect or exploit short-term stock activity: creating false breakouts, pushing prices through clustered stop levels, trading aggressively near the close, and repeatedly capturing small spreads. It…
This research summary argues that broker stocks deserve separate modeling within CSI 300 enhancement portfolios because the sector has a meaningful index weight and tends to move differently from other industries. It tests stock-selection factors within the…
The document describes a Force indicator calculated from trading volume and the change in a moving average of closing prices. Its inputs are close, volume, and a configurable period. The indicator multiplies volume by the difference between the current…
This Chinese-language competition update reports diagnostics for the ten highest-ranked anonymous factors and tracks a CSI 1000 index-enhancement portfolio. It describes average style exposures estimated through daily multivariate regressions and…
The document asks how to express a proposed factor in a platform's formula language. The factor is defined as the negative correlation, over a rolling interval, between a security's VWAP series and its volume series. In principle, this transforms the…