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
This stock-selection proposal combines a 14-period RSI threshold below 65 with circulating market capitalization. It includes stocks in a stated mid-cap range as well as those above that range, then ranks qualifying names by price change and selects up to…
This notebook demonstrates a pyfolio workflow for examining one stock’s returns against the canonical Fama–French factors. It first plots rolling factor betas directly from the stock return series, then calculates those betas for use as benchmark returns in…
This A-share selection method looks for turnover between 3% and 12%, year-over-year growth in net profit attributable to parent-company shareholders above 20% and no more than 100%, non-ST status, and a five-day rising-price condition described as a limit-up…
This article explains how to allocate weights in a multi-factor equity portfolio using a structured risk model. It decomposes stock returns into factor exposures, factor returns, and stock-specific returns, then describes industry and style neutrality…
The article proposes a Chinese equity screen combining a large intraday price range, a high annual dividend measure, and elevated relative trading volume. The stated selection rules require amplitude above one, a 2019 dividend proportion above 25%, and a…
This document describes a Chinese stock-selection screen using three conditions: a 14-period RSI below 65, an outside-volume to inside-volume ratio above 1.3, and year-over-year parent-company net profit growth above 20% and no greater than 100%. The…
This research note describes a China A-share fund designed to enhance returns over the CSI 500 while managing benchmark-relative risk. It attributes the strategy’s excess returns to quantitative multifactor stock selection, sector allocation, trading…
This selection rule combines a price-range filter with a market-universe restriction and a profitability screen. It targets stocks whose codes begin with 60, whose amplitude exceeds a threshold, and whose return on equity has remained above a stated level…
This Chinese-language post describes a stock screen that combines daily price amplitude, a historical dividend measure, and the day’s change in reported holdings. The rationale is to pair stocks with noticeable price movement and a high dividend measure with…
The discussion addresses how to search for effective weights when combining two stock-ranking factors. It proposes using BigQuant’s parallel tuning module, m.tune, rather than treating each weight as a separately exposed parameter. The key implementation…
This stock selection method combines a price-range condition, a ratio involving yesterday’s turnover and today’s auction volume relative to yesterday’s volume, and historical revenue growth. The stated screen requires the turnover-volume ratio to fall…
The document outlines a way to construct daily, cross-sectional high-frequency equity factors. It proposes ranking stocks by their volume as a percentage across the market, then calculating the cross-sectional variance, skewness, and kurtosis of those ranks.…
This strategy note outlines a Chinese equity model that ranks stocks using three cross-sectional features: market capitalization, turnover, and five-day price change. Each feature is converted to a daily percentile rank. A shallow XGBoost regressor is…
This analysis examines weakening one-month return reversal and specificity factors in Chinese equities. It argues that judging a long-short portfolio alone can conceal deterioration: although aggregate reversal returns remained positive, the long leg had…
This Chinese equity-selection idea combines three screens: a 14-period RSI below 65, year-over-year growth in parent-company net profit above 20% and no more than 100%, and money-flow strength ranked from high to low. The example implementation estimates…
This strategy combines a qualitative earnings-surprise screen with the stock’s price gap after an earnings announcement. It identifies surprise candidates by analyzing analyst report titles for language suggesting results exceeded expectations, rather than…
This document outlines an equity screening rule that combines daily price amplitude, a measure labeled today’s controlling shares, and company capitalization. The screen selects stocks when amplitude exceeds 1, the controlling-share measure exceeds 21, and…
The post records a China A-share backtest using market capitalization ranks to select smaller companies, with filters for trading activity, positive trailing earnings, and positive price-to-earnings ratios. It reports that increasing the portfolio from 10 to…
The article reviews a research method that assigns firms to startup, growth, maturity, shakeout, or decline stages using the signs of operating, investing, and financing cash flows. The three cash flow signs create eight possible patterns, which the method…
This research summary compares machine-learning stock-selection models trained on conventional return labels with models trained on alternative labels based on information ratio and Calmar ratio. The rationale is that these labels capture more of the path…
This Chinese-language post presents an equity screening idea that combines daily amplitude above a threshold, a positive weekly MACD reading, and a condition described as yesterday’s main-force control. Its proposed refined screen adds valuation and income…
The document describes a method for converting company-level factor views into industry-level indicators by aggregating the factors of each industry’s constituent stocks. It reports that mapped measures based on changes in asset turnover and leverage,…
The document describes a framework for separating active fixed-income fund returns into persistent strategic factor exposure, short-term tactical exposure, and residual return, which it treats mainly as security selection. Long-run factor betas are estimated…
This post starts with a proposed stock screen based on amplitude above 1, a control-related measure above 21, and positive return. Its expanded version adds relative performance against a benchmark over roughly three months and valuation and profitability…