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.
21,023 documents
This Chinese stock screen combines three conditions: daily price amplitude above 1, turnover between 2% and 9%, and a ten-day price gain greater than zero but below 35%. The stated rationale is to find shares with meaningful price movement and trading…
This brief indicator description presents a classic exponential moving average for MetaTrader 5, with support for real-valued averaging periods. It distinguishes two versions: EMA_.mq5 can be applied to other indicators, while EMA.mq5 lacks that capability.…
The document explains a MetaTrader function that identifies historical price peaks and troughs using a minimum price-change threshold. It describes the method as similar to the FastZZ indicator: the user supplies quote history and a required move size, and…
This Chinese equity screening proposal filters stocks using a turnover range, a limit on circulating shares, and a minimum total share count. Its stated criteria are turnover from 3% to 12%, circulating shares no greater than 5.5 billion, and total shares of…
This proposed stock screen combines convergence among at least five moving averages with more than two limit-up days in a ten-day window, framed in the document around 2021 data. The rationale is that closely grouped averages may indicate relatively stable…
The document describes a script that applies empirical mode decomposition to price data for the currently selected trading symbol and displays the output in a web browser. It identifies the method by name and refers readers to a separate introductory article…
The document outlines a class for building a generalized regression neural network. Its constructor sets input and output vector sizes, while a learning method accepts training patterns arranged in flat input and output arrays. Training stops when it reaches…
The document describes a class for calculating trading statistics from trade history, including while a strategy is running or outside the strategy tester. It presents this as a way to access statistics without relying on the tester-only statistics function.…
This document presents a screen that ranks stocks by volume ratio, selects the top 100, requires a positive price-to-earnings ratio, and checks that the 30-day moving average is rising. It frames relative trading activity and an upward short-term trend as…
This document outlines a simple Chinese equity screen that ranks stocks by money strength, limits share price to below 12, and applies a 2021 time filter. Its stated rationale is to favor lower-priced shares with stronger money-flow readings and to identify…
This document explains a market-regime indicator built on Bag-of-SFA-Symbols (BOSS). It converts normalized price windows into symbolic words using a low-pass Fourier representation and data-learned bins, then compares stretches of market history as…
The document proposes a China stock screen that ranks companies by capital strength, using measures such as turnover or trading value, and looks for a weekly MA5 crossover above MA10. Its initial concept also calls for market capitalization below 10 billion…
This note describes a fundamental screen based on company market capitalization and avoiding firms with losses. It also recommends considering profitability, growth, industry trends, and policy conditions when judging investment value. A short Python example…
The article proposes ranking stocks by fund-strength measures and filtering for companies whose 2021 revenue was more than 1.1 times their 2018 revenue. It presents fund inflows as a possible indicator of market interest and the revenue comparison as a…
The document presents a stock-ranking idea framed around a morning-star pattern and a named stock, but its practical selection logic is primarily to order stocks by a capital-strength score. The suggested score divides a ten-period moving average of closing…
The Aggressiveness indicator measures an instrument’s average movement in points per candlestick over a configurable number of recent periods. It counts movement regardless of direction, so it describes typical candle movement rather than whether price is…
Tension Flow Trend combines a Hull moving average baseline with a Z-Score of price’s deviation from that baseline. The baseline slope sets the directional colour, while the deviation divided by its rolling standard deviation measures how far price has…
The document describes an expanded version of the Awesome Oscillator, a momentum indicator available in MetaTrader 5. It retains the simple moving average as the default calculation while allowing exponential, smoothed, or linear weighted averages. Users can…
This essay outlines common reasons a promising backtest may fail in live trading. It recommends checking calculations when returns look implausibly strong, including comparing the account’s initial and current states to derive profit and loss. It also warns…
The document explains quicksort, a recursive method for sorting an array by selecting a pivot and partitioning values around it. It outlines a partition process using left and right pointers, swapping out-of-order values until the pointers meet, then…
The Coppock indicator combines two rates of change, typically measured over 14 and 11 months, then smooths their sum with an exponential moving average, commonly over 10 months. The document traces its origin to E. Coppock in the 1960s and describes its…
This post describes an equity screen that combines price movement and recent strength: amplitude above one, a rising price bottom, and at least one daily gain of 10% or more during the previous 25 trading sessions. It presents the conditions as a way to find…
This document specifies a screen for Chinese main-board stocks with turnover between three and twelve percent, floating market capitalization between 5 and 10 billion yuan, and a closing price below 12 yuan. It includes a formula-style condition and a Python…
This Chinese-language account describes a learning project that builds a China stock selection strategy with five ranked factors and a two-day forward price label. It trains a Ridge regression model, standardizes the inputs, and periodically retrains before…