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.
4,194 documents
The post proposes a stock screen combining a daily high-low range threshold, a history of strong return on equity, and a newly formed KDJ turning signal. It interprets the range condition as a volatility filter, the profitability condition as a quality…
The document explains alpha as a signal or mathematical expression that can be converted into portfolio weights, then outlines a research cycle from investment hypothesis and data choice through backtesting and review. It discusses potential signal sources,…
This project report describes using a model with the BigQuant dai function’s syntax and formulas as context, then prompting it to build increasingly complex quantitative factors. The author says ordinary prompts and iterative adjustment moved the generated…
This stock-selection note proposes screening companies in the beverage and alcohol import-export industry using an RSI below 65 and a 2019 dividend payout ratio above 25%. It presents RSI as a measure of price conditions and the historical payout ratio as a…
This notebook illustrates factor evaluation with Alphalens using a large-cap equity universe assigned to sectors. It compares a baseline factor based on each stock’s recent ten-day performance with a second factor constructed from future price changes. The…
This notebook demonstrates how to use Alphalens to compare a deliberately non-predictive factor with a deliberately predictive one. It uses a universe of large-cap stocks with sector labels and daily opening prices. The baseline factor ranks stocks by their…
This stock screen selects shares with a daily high-to-low range above one percent, excludes Beijing-listed A shares, and rejects stocks that reached the upper price limit on the previous day. The document presents short-term volatility as a way to find…
This stock selection idea screens companies in the metaverse theme for circulating market capitalization above the stated threshold and a price-to-earnings ratio below the stated ceiling. The accompanying discussion presents the screen as a simple…
This stock screen combines a technical condition with reported earnings growth and a price-shape filter. It selects A-shares with a 14-period RSI below 65, parent-company net profit growth from above 20% through 100%, positive net profit, and a…
This short forum post considers whether combining quantitative stock selection with discretionary choices improves results. The author argues that adding human stock picks to a quantitative selection process may perform worse than either approach alone, and…
This study turns Chinese analyst-report summaries into equity signals using a fine-tuned BERT sentiment model. It starts with a pretrained Chinese model, adapts it on labeled financial opinion text, and applies it to unlabeled report summaries.…
The article examines how large datasets and complex models can make factor discovery more productive while increasing the risk of overfitting. It contrasts hypothesis-led research, where an economic or behavioral rationale is tested, with data-led mining,…
This stock screen combines a daily amplitude threshold, a positive weekly MACD-style signal, and moving-average alignment with valuation and dividend filters. Its final stated conditions require amplitude above 1, a weekly positive bar, upward separation of…
This equity research summary describes high-frequency factors built from combinations of price and trading-volume information. It presents two examples: an illiquidity factor adjusted using changes in the price path, and a trading-behavior factor derived…
This research note proposes a way to study the microstructure behind equity reversal. For each stock, it examines the prior twenty trading days, calculates average transaction value per trade for each day, and separates days with relatively high and low…
This research summary argues that factor returns can vary nonlinearly and across groups of stocks, so a single linear factor relationship may miss meaningful differences. It contrasts direct nonlinear transformations, which can be hard to justify…
This post outlines a Chinese A-share screen combining three inputs: RSI below 65, year-over-year growth in net profit attributable to the parent company above 20% and no more than 100%, and positive net buying attributed to major participants during the…
This stock-selection screen combines a technical condition with two fundamental filters. It selects A-shares whose RSI is below 65, whose year-over-year growth in net profit attributable to parent-company shareholders is above 20% and no greater than 100%,…
This post describes a Chinese equity screen combining an intraday high-low range above 1, return on equity above 15% for five consecutive years, and prior-day auction turnover above 0.26. The intended mix is a price-activity condition, a multiyear…
This research summary compares minute-data factors across equities and futures, grouping them into return-distribution, volume-distribution, combined price-volume, order-flow, and trend-strength measures. It reports that several factors behave differently by…
This research roundup summarizes two studies. The first revisits the equity size effect, the long-standing claim that smaller-capitalization stocks earn higher average returns than larger ones. It describes the study as examining competing claims about the…
This factor proposal estimates intraday active buying from minute bars: when a bar closes above its open, its traded amount is counted as active buying; bars closing below their open are assigned to passive buying. The daily active-buy ratio divides the…
The report summary presents a Campisi-style performance attribution model for bond funds using net asset value returns, an approach intended for cases where timely holdings data are scarce, especially for private funds. It describes five explanatory factors:…
The document proposes a Chinese equity screening rule for Shenzhen main-board stocks in the metaverse theme. It combines positive institutional-flow readings with positive price-to-earnings ratios below 29.01 and price-to-book ratios below 3.11. It also…