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 document describes a MetaTrader indicator that measures how quickly quotes arrive relative to the normal pace for that time of day. It converts quote changes into events, combines records stamped in the same millisecond, treats long gaps as session…
This trading system uses the GG-RSI-CCI indicator to trigger trades when a bar closes and its color changes from red to green or blue. Open positions are closed either by a pending order or when the indicator’s bar color changes. The document does not…
This excerpt introduces a rule-based strategy built around the Ichimoku indicator and begins a buy condition: price must be above the Kumo cloud. The author describes the approach as suitable for automation and reports personal observations of favorable…
This stock selection idea screens for shares with turnover between 3% and 12%, a sequence of seven declining closes, and no limit-up event on the previous day. The post provides both a formula-style reference and a Python outline for checking turnover,…
This study applies TabNet, a neural network architecture for tabular data, to A-share stock selection. It outlines TabNet's encoder structure, in which attentive masks select features at successive steps, and explains the motivation for combining neural…
This Chinese-equity screening rule combines a turnover band, proximity of the opening price to a short moving average, and a longer-term trend condition. It selects stocks with turnover from 3% through 12%, an open within 5% of the 10-day moving average, and…
This example describes a daily AI-driven process for selecting sector ETFs. Five agents cover technology and communications, financials, healthcare, energy, and consumer sectors, each proposing an idea. A risk manager reviews the pitches for crowded…
This report summary describes a forecast of additions to and removals from the CSI 300 index ahead of a scheduled constituent review. It says the review follows published selection rules, including a limit on the share of constituents that can change, buffer…
This documentation overview maps backtesting data sources to asset classes, bar intervals, and setup requirements in LumiBot. It suggests Yahoo for uncomplicated daily stock and ETF tests, ThetaData for intraday stock or options history, Polygon for several…
This Chinese equities screening proposal selects stocks using three conditions: daily amplitude above 1, a positive weekly MACD histogram, and turnover between 3% and 12%. The rationale is that price range and a positive weekly momentum signal may indicate…
This example demonstrates how to model an option purchase across expiration in a backtest. A strategy subscribes to option quotes and futures bars, then submits a market buy order for one option when it receives the first eligible quote. The sample replays…
The document presents a market-depth implementation that stores bid and ask quantities by integer price ticks in ordered B-tree maps. It tracks the best bid and ask, converts prices to ticks using a configured tick size, and filters near-zero quantities…
This research summary proposes measuring stock valuation through time as well as across stocks. It separates a valuation series into a longer-run trend and a short-term deviation, then uses an error-correction model to define deviation factors from the…
The document gives a broad introduction to quantitative trading as the use of data, statistical or machine-learning models, and programmed rules to inform and execute investment decisions. It surveys common components: historical and live market data,…
This module describes calendar rules for rolling several futures series: crude oil, NBP natural gas, refined products including RBOB, grains, and ethanol. The rules use contract-specific termination conventions, such as dates near the 25th or 15th of a…
The article surveys how market and blockchain data may inform long-term crypto investing and short-term trading. For fundamental research, it lists measures such as market capitalization, supply, trading activity, network use, token holders, velocity, total…
This Chinese A-share screening idea combines a 14-period RSI threshold with listing age and recent limit-up activity. It seeks stocks with RSI at or below 65, more than one year since listing, and at least two limit-up events within a 500-day window. The…
This document describes an Expert Advisor that trades from signals produced by the iWPRSign indicator. A signal is acted on when a colored indicator icon appears at the close of a bar. The system depends on a separately compiled indicator file, so it cannot…
This forum post raises a backtesting problem: a simple EMA20 strategy on TQQQ reports that account funds fell to zero or below, preventing the platform from calculating performance statistics. The trader questions the warning because the listed trade entries…
The document describes a MetaTrader 5 library for testing expert advisors against simplified proprietary trading challenge rules. It records an initial balance, calculates profit or loss relative to that amount, and uses tester deposit or withdrawal…
The document describes an audit of terminal-held price bars before they are used for indicators or backtests. It checks OHLC consistency, bar timestamps, missing openings relative to the broker’s stated quote sessions, empty or flat bars, higher-timeframe…
This article explains cross-sectional momentum in commodity futures: periodically rank markets by recent performance, buy the strongest and sell the weakest. It distinguishes this relative ranking approach from time-series momentum, which evaluates each…
This article outlines a machine-learning stock-selection experiment using TabNet and 98 price-and-volume factors. It trains the model on data from 2010 through 2018, then applies its predictions in a backtest covering 2018 through September 2021. The…
This tutorial outlines a machine-learning workflow for selecting Chinese A-share stocks with a convolutional neural network. It describes defining a stock universe and input factors, extracting and splitting data, turning observations into rolling sequences,…