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 Expert Advisor combines two smoothed XPeriodCandle indicators across different timeframes. A slower indicator sets the prevailing trend from the color of its most recently closed candle, while a faster indicator provides entries when its candle color…
The document describes an Expert Advisor intended for five-minute EUR/USD trading. Its author says the design aims to collect small profits steadily while sometimes benefiting from a strong trend, and points readers to supplied settings for further…
This community post discusses a learner's flowcharts for VeighNa, an algorithmic trading framework, and clarifies how a bar callback works. The described process aggregates tick data into a one-minute bar. When the tick time moves into a new minute, the…
This report describes a stock-selection approach that optimizes portfolio weights while constraining market-cap, industry, and style exposures. Its goal is to retain exposure to alpha factors while limiting unwanted variation from other risk factors. The…
This document describes an expert advisor that trades when a smoothed Heiken Ashi indicator changes candle color. It checks for the change at bar close and opens a trade in response, using the transition between blue and red as the signal. The system depends…
This Chinese-language announcement outlines an advanced course on developing machine learning cross-sectional multi-factor strategies with VeighNa. Its subject matter includes factor data preparation, feature cleaning, target selection, model training and…
This notebook walks through a minimal Qlib reinforcement learning setup, connecting a simulator, state and action interpreters, a reward function, a policy, a dataset, and training and backtest workflows. Its simulator runs for a fixed number of steps,…
This Expert Advisor uses a Stochastic Histogram indicator and evaluates its signals when a bar closes. A selectable trend mode determines how signals are recognized. One method responds to a crossing between the indicator’s main and signal lines together…
The document combines a strategy tester report with a description of requested rules for an Expert Advisor using three indicators. The proposed sequence uses a faster and slower linear weighted moving average crossover as the initial direction signal, an…
The article describes a Chinese equities screen that selects stocks from a 2021 period based on a reported increase in position share above 5% and a daily gain greater than 1. It interprets the first condition as a possible sign of buying interest and the…
This guide explains that a strategy's execution mode is determined by the runner call and its data or broker configuration. Calling the class's backtest method starts a historical simulation, while constructing it with a broker and invoking the live runner…
This Chinese community post proposes a stock-selection screen for companies classified in the metaverse sector. It selects names that had at least one limit-up session within the previous 25 trading days and whose turnover rate averaged between 2% and 9%…
This tutorial section explains how to implement a cross-sectional momentum strategy, commonly called Up Minus Down, in Moonshot, an open-source vectorized backtester that uses pandas. The example ranks securities by returns over a 252-day momentum window…
This example describes a daily rule that uses CNN’s Fear & Greed Index to set exposure to SPY. A research agent retrieves a recent score, and a trading agent assigns one of five SPY allocations: full exposure at extreme fear, progressively smaller…
This review examines whether published indicators of future mutual-fund performance retain their predictive power outside the periods in which they were studied. It compares high-minus-low portfolio alpha spreads and cross-sectional regression slopes across…
This Qlib documentation explains how forecast models produce stock prediction scores and how to train and run a model independently or within an automated workflow. It describes the base model interfaces, including support for fine-tuning, and gives a…
This gold strategy uses a turn in the stochastic signal line to indicate a possible momentum shift, then waits for price to cross the previous day's high or low before entering with a stop order. It calculates its stochastic variant internally, using a…
The document describes an Expert Advisor that trades signals from the Iin_MA_Signal indicator. A trade signal is recognized when the indicator displays a colored icon at the close of a bar. The EA requires the compiled indicator file to be installed in the…
This article translates an investor’s concentrated value approach into a quantitative stock-selection framework. It seeks companies with reasonable valuations, durable operating performance, substantial insider ownership, and share prices below their recent…
This document reports a two-year daily Bitcoin–USDT backtest on Binance, covering January 2019 through January 2021. The strategy is identified only as “AwesomeStrategy”; no entry or exit rules, position sizing, or execution assumptions are supplied, so the…
This note describes a Chinese-equity screen that combines a price-amplitude threshold, simultaneous bullish crosses among three moving averages, and exclusion of stocks that hit the prior day's upper price limit. Its example uses short, medium, and longer…
This Chinese-language post outlines a stock-selection screen based on turnover of 3% to 12%, a positive return over ten days below 35%, and exclusion of stocks marked ST. It proposes running the selection before 10 a.m. and combining the filters with a named…
This stock-selection rule combines elevated amplitude, a rounded price pattern, and at least one daily gain of 10% or more during the preceding 25 trading days. The accompanying formula adds more precise conditions: it checks recent highs and lows, requires…
This proposed daily premarket stock screen selects shares with MACD above zero, a positive price-to-earnings ratio, and a closing price equal to 18.5 yuan. The article interprets positive MACD as an uptrend and positive P/E as a normal valuation measure. It…