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.
16,761 documents
This assignment response translates two discretionary stock approaches into rule-based proposals. One combines recent institutional fund inflows, positive company earnings, improving per-share profit, elevated trading volume, and a price ceiling relative to…
This forum post describes an AttributeError in a BigQuant high-frequency backtest. The copied trade-module code treats each key in the portfolio positions mapping as an object with a symbol attribute. In the HFTrade interface, the key is already a string…
This research summary examines quantitative stock selection among Chinese technology companies. It highlights research and development spending as a candidate signal and also discusses profitability, earnings growth, valuation, company size, turnover, and…
This example builds a simple portfolio analysis workflow that generates a daily value series for several allocation weights and plots the paths together. A configuration object holds the tested weights, chart dimensions, and date range. The demonstration's…
This project explores combining strategies associated with different market styles. The author says market styles can persist over a period, so a strategy that fits a clearly expressed style may adapt better to prevailing conditions. They changed a provided…
This research summary examines analyst recoverage: the first new recommendation after an analyst or brokerage has stopped covering a stock for at least six months. It compares recoverage with initial coverage and ordinary rating changes, using U.S. analyst…
The document describes a beginner’s question about passing results from earlier BigQuant modules into a backtest. The proposed strategy uses a fixed universe of ten stocks, ranks them daily by five-day return in ascending order, buys the five lowest-ranked…
This Chinese equities screen selects stocks with turnover between 3% and 12%, circulating share capital no greater than 5.5 billion shares, and at least two limit-up sessions in a rolling 500-day window. The stated idea is to combine trading activity and a…
This document describes an Expert Advisor that trades signals from the BinaryWave_StDev indicator. It opens a position when a colored indicator point appears at a bar’s close, then exits when the indicator reverses direction against the open position. The…
This indicator plots historical average returns for each calendar month using monthly price data. After a user selects a start month and year, it groups each month’s open-to-close changes across the available history and averages them. It displays the twelve…
This Chinese equity screening proposal combines three conditions: turnover between 3% and 12%, three consecutive declining closes, and a rising DEA line from MACD. The stated rationale is to look for stocks with potential upside after a short run of losses,…
This research reproduction describes factors based on the share of a Chinese stock’s trading volume occurring in the opening call auction and near the close. The opening-auction factor, OCVP, is smoothed with a simple moving average and an exponentially…
This Chinese-language research summary studies whether public equity fund stock exposure can inform market timing in the China A-share market. It uses a moving-average system to distinguish trending from range-bound regimes, analyzes how fund positioning…
The document introduces VeighNa, an open-source Python framework for quantitative trading, with particular attention to its vnpy.alpha module. That module organizes research into feature creation, model training, strategy development, and workflow…
This guide presents a relative strength index strategy using overbought and oversold thresholds. It describes calculating RSI from rolling average gains and losses, generating short signals above 70 and long signals below 30, and optionally filtering trades…
This research report proposes improving a conventional stock reversal signal by splitting each stock’s recent daily returns according to average trade size. For each lookback window, it ranks days by daily turnover divided by trade count, compounds returns…
The document describes a three-candle Popgun pattern: an inside candle is enclosed by two outside candles. It uses the direction of the second outside candle to choose long or short bias, then treats a break beyond that candle’s high or low within the next…
This tutorial demonstrates a graph convolutional policy, GPM, inside a reinforcement-learning portfolio workflow. It loads historical stock features and a sector and industry graph, then reduces the graph to nodes within two hops of the selected portfolio…
The document outlines a Chinese A-share stock screen centered on a metaverse industry classification, elevated recent turnover, share prices above a stated threshold, larger market capitalization, and return on equity. It first presents a simpler theme,…
This post outlines a Chinese equity selection idea that ranks stocks by capital-flow strength, choosing the top 100, and filters for an opening-stage price rise below 6% at 9:25. It associates strong flow rankings with market attention and treats a limited…
This post describes a short-term Chinese equity screen using RSI below 65, an outside-volume to inside-volume ratio above 1.3, and a share-price filter. Although the headline mentions a specific price, the body shifts to a range: its example uses prices from…
This post proposes screening Chinese equities for daily price amplitude above 1, return on equity above 15% for five consecutive years, and a rounded price pattern. It presents the combination as a way to pair a volatility condition with sustained…
This BigQuant platform report investigates Beijing Stock Exchange records in the Chinese stock factors table and how they interact with a basic stock-selection query. The author queries instruments with the Beijing suffix for a single date and reports 249…
This indicator description presents a TRIX variation that uses Wilder-style double exponential smoothing and floating signal levels. It offers three possible ways to mark changes on a chart: crossing the outer levels, crossing a dynamic middle or zero level,…