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.
79,386 documents
The strategy models a refining spread using crude oil, fuel oil, and a third petroleum product in a 3:2:1 weighting. It calculates the spread as the weighted value of the two product legs minus the weighted crude leg, then compares the current spread with…
LeManChanel is described as a technical indicator that estimates the possible boundaries and direction of the next bar. The document identifies the indicator’s original implementation as an MQL4 publication from 2009 and credits LeMan as its author. The…
This document describes an equity screen combining a 3%–12% turnover-rate range, a candlestick reversal condition, and a circulating-share limit of 5.5 billion shares. It gives a corresponding formula and sample data workflow that calculates a reversal…
The document describes a chart indicator that displays bull power and bear power together in one MetaTrader 5 window. Its stated motivation is that the two measures have different value ranges, which makes a direct overlay in a shared scale difficult to…
This article presents a method for identifying time windows in which an existing trading system performs better. A library analyzes the system’s trading history, removes weaker intervals, and reports results for the remaining periods. In its scalper example,…
This post describes a Chinese equity screening rule using three conditions: daily high-low range above 1% of the prior close, a positive but less than 35% return over ten days, and at least one limit-up event during the prior 25 days. The stated rationale is…
This document presents a QuantStats tear sheet for a strategy labeled “buffett-plain,” compared with SPY over January 4–15, 2026. It lists returns, drawdowns, risk-adjusted statistics, market exposure, daily outcomes, and two drawdown episodes. The reported…
This tutorial demonstrates a component-level backtest workflow using NautilusTrader. It loads historical Binance ETH/USDT trade ticks, configures a simulated spot venue with a cash account and maker-taker fees, and aggregates ticks into bars. A strategy…
This brief indicator description introduces a trend tool that displays a colored histogram based on the difference between a Hull Moving Average and an averaging of that moving average. Its stated purpose is to make the trend signal visible in histogram…
This report compares hourly Bitcoin returns, volatility, and trading volume across Binance, Bybit, Coinbase, HTX, Kraken, and OKX. It groups activity by the trading hours of several regions, using standard deviation of hourly returns as a volatility proxy…
The post addresses missing final one-minute bars when a closing or session-boundary tick is not delivered. Its proposed fix adds logic to a bar generator’s tick-update handler: near selected minute boundaries, it checks for a tick arriving in the final…
This Chinese-language post proposes a stock screen combining three conditions: daily amplitude above a threshold, today’s price relative to moving averages in an upward configuration, and a prior-session signal described as main-force control. The rationale…
The document describes an accumulated version of the Relative Strength Index (RSI). It forms a new oscillator by combining consecutive RSI calculations, then interprets that oscillator with adaptive thresholds such as floating levels or quantile bands. The…
The document describes an indicator that displays the current trend using values from the Skyscraper_Fix indicator on a fixed timeframe. The signal is derived from the color of the original indicator’s middle line. Markers appear on the chart when a bar…
The discussion raises two practical issues for CTA strategies: protecting source code when running a strategy on another computer or server, and adapting a strategy as a futures market’s active contract changes. Replies point to a strategy-encryption guide…
The discussion clarifies that maximum drawdown duration measures how long the strategy remains in its largest drawdown. It distinguishes the time span of a decline and recovery from the size of the loss itself, which is what the maximum drawdown percentage…
This post outlines a rule-based screen for Shanghai-listed shares with codes beginning in 60. It first restricts candidates to stocks with turnover between 3% and 12%, then requires the previous day’s price to exceed its 250-day moving average. The stated…
This weekly competition report presents diagnostic summaries for the ten highest-ranked submitted factors and tracks a China 1000 index-enhancement portfolio. Factor diagnostics include average exposures to BARRA style factors and sector-level average Rank…
This introductory note describes quantitative investing as turning research, asset selection, timing, and position decisions into explicit rules that can be tested and executed by a computer. It distinguishes cross-sectional stock selection, which compares…
The article argues that a sharp selloff in World Liberty Financial Token (WLFI) preceded a broader cryptocurrency decline and may have offered an early warning of market stress. It highlights a reported surge in WLFI trading volume after tariff news,…
This introductory overview explains quantitative trading as the conversion of market observations and decision rules into mathematical models and computer programs. It outlines a typical process: analyze historical price, volume, and timing data; design a…
The document describes a Chinese A-share stock screen focused on the metaverse sector. It selects stocks whose opening price is less than 6% above the prior close, whose price chart resembles a rounded arc, and whose Bollinger Band width is below 0.05. The…
This overview introduces reinforcement learning as a way to learn sequential decisions by interacting with an environment and maximizing accumulated reward. It outlines the agent, environment, policy, and reward, and explains how delayed feedback differs…
This indicator description explains how to display JMACandle bars from a higher timeframe on a chart. It draws each candle as a filled rectangle using fill buffers, distinguishing the candle body from its shadows with separate colors. Up and down candle…