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.
29,558 documents
This Chinese equity screen selects stocks with turnover between 3% and 12%, a circulating share base no larger than 5.5 billion shares, and positive or high-ranking large order net activity. The stated rationale is to find stocks with moderate trading…
This stock selection approach combines technical, flow, and company filters. It looks for stocks whose 5, 10, 20, 60, and 120 day moving averages cluster near the same price, which the article treats as a potential buy signal. It then favors positive large…
This note describes a daily stock screen combining three conditions: the stock code begins with 60, the session’s high-to-low range exceeds 1% of its open, and the close is below the open but above the prior close. The last condition identifies a bearish…
This example configures a backtest for a mean-reversion strategy on an AUDUSD perpetual contract. It feeds quote data into a backtest engine, forms one-minute midpoint bars, and instantiates a strategy configured with Bollinger Bands and RSI. The listed…
The document describes a Chinese equity screen combining three conditions: amplitude above a threshold, a ranking by large-order net volume, and a three-day limit-up streak. It presents these as ways to find active stocks with recent price strength. The…
This proposed Chinese equity screen selects stocks using turnover, circulating share size, and recent limit-up activity. Its initial conditions call for turnover between 3% and 12%, a circulating share count no greater than 5.5 billion, and more than two…
This note explains an automatic envelope channel drawn parallel to a moving average. The channel width is calibrated so that it contains roughly 95% of prices observed over the prior two or three months on a daily chart, leaving only unusually high or low…
This Chinese stock screen selects shares with amplitude above 1, a closing price above the five-day moving average, and a positive return. Its indicator examples express amplitude relative to the previous close, compare the close with the moving average, and…
The document describes the cup and handle as both a possible reversal pattern after a decline and a continuation pattern within an existing uptrend. A rounded cup is followed by a pause or pullback called the handle. The proposed long entry occurs when price…
This Chinese-language post describes a stock selection rule combining three signals: daily high-low range above 1%, a newly formed upward KDJ crossover, and a 14-period ratio of volume on rising sessions to volume on falling sessions above 1.3. It proposes…
This tutorial explains a two-input market-making setup for a Lighter perpetual linked to Nvidia shares. The Lighter order book supplies the price anchor, while Databento US equity top-of-book quotes provide a normalized signal: the equity mid is compared…
This stock-selection idea combines daily amplitude above 1, a positive weekly MACD histogram, and return on equity above 15% for five consecutive years. It pairs a price-activity condition and a technical momentum signal with a measure of sustained…
This Chinese stock-screening post combines four conditions: amplitude above 1, current trading volume above 10,000 lots, an open above the previous close, and a 15-minute MACD histogram described as shortening while negative. It presents these as signs of…
This brief description outlines an expert advisor that builds a series of trades in one direction. The direction is set by the first position, and subsequent positions are opened at what the description calls the best price while increasing trade volume. It…
This Chinese stock-screening example selects shares in the metaverse theme that close above their 250-day moving average and have recorded at least two limit-up sessions over the past 500 days. The author frames the long moving average as a trend or…
This document describes an Expert Advisor that predicts a future price from a sequence of past prices using linear prediction. Burg’s method estimates model coefficients by reducing mean-square prediction error over a training window. The strategy’s…
This stock screen combines three conditions: turnover within a stated range, appearance on the prior day’s market leaderboard, and positive net buying by major participants during the opening auction. The article interprets these filters as a way to find…
The document introduces quantitative trading as the use of computers, mathematics, statistics, and a systematic process to develop signals for buying and selling. It briefly describes the field’s history, including Jules Regnault’s work on price variation,…
This code provides several ways to assess how a backtested equity portfolio might interact with market liquidity. It aggregates executed shares by ticker and day, compares those totals with daily bar volume, and identifies each name’s largest observed share…
This guide explains how pre-trade controls can block orders that exceed preset limits, helping reduce accidental oversizing, excessive order flow, and other operational errors. It describes common controls for order frequency, reset intervals, single-order…
This example demonstrates how a backtest engine can model automatic liquidation on a margin account holding an inverse Bitcoin perpetual. It configures a simulated venue with liquidation enabled, starts with one BTC, and submits a market buy for 10,000,000…
The strategy uses the close at a chosen hour and weekday as a reference price. It buys or sells when price moves beyond a user-set fraction of the average daily range, with stop-loss and take-profit distances also defined as fractions of that range. The…
This expert advisor combines Heikin-Ashi color changes with a Keltner-style channel to time reversal entries. A color flip arms a signal, but entry waits for price to close back into or through the channel in the new direction. Signals expire after a…
This document describes a forex trading method built around a CCI oscillator smoothed with a T3-style calculation. The indicator is said to generate directional signals as it moves across the zero line. The suggested entry waits for the second red or green…