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
This document introduces HftBacktest, a Rust framework for developing and running high-frequency trading and market-making strategies. Its backtesting approach replays tick-level market data and aims to model important execution effects, including feed…
The post presents a stock screen based on three conditions: today’s position-growth share exceeds 5%, the close is above the previous day’s low, and the 10-day return is positive but below 35%. It frames these filters as a way to find stocks with recent…
This Expert Advisor tests a time-based breakout approach. At 15:00 Moscow time, it places a buy stop and a sell stop 100 points from the current price, with a 200-pip stop loss on each order and no take-profit target. If one order triggers, the other remains…
This A-share screening proposal combines a 2019 dividend payout ratio above 25%, a daily gain above 1% for main-board stocks, and reported buying activity above 5%. The article interprets the dividend measure as a possible sign of business strength and the…
This Chinese equity screening proposal combines an intraday high-low range of at least 1%, a nonempty name for an outstanding convertible bond, and positive net buying attributed to major participants during the opening auction. The accompanying indicator…
This proposed Chinese stock screen selects companies associated with the metaverse theme, with a 14-period RSI below 65 and a positive 10-day return under 35%. The article frames RSI as a short-term sentiment gauge and the theme filter as a way to target a…
This A-share screening idea combines three signals: a reported increase in buying activity above 5%, year-over-year parent-company net profit growth above 20% and no more than 100%, and a positive 10-day return below 35%. The article then proposes adding…
This Freqtrade strategy defines long entries on a five-minute chart using MACD, RSI, and Bollinger Bands. It requires MACD to be above zero and its signal line, the upper Bollinger Band to be rising, and RSI to exceed 70. Long exits are triggered when RSI…
The article explains an RSI divergence indicator that avoids confirmed pivots. Instead, it compares the current price and RSI with extremes from an earlier short window, then requires elevated volume and an RSI extreme before marking a bullish or bearish…
This report reviews 2024 activity in centralized crypto spot and derivatives markets. It describes how exchange and asset trading volumes changed through the year, noting periods of high activity across major venues and tokens. It suggests using volume…
This proposed A-share screen looks for stocks with RSI below 65, an external-to-internal trading volume ratio above 1.3, and no limit-up close on the previous day. The article frames the RSI threshold as a way to avoid heavily overbought stocks, treats the…
This stock screen combines a daily price range threshold with a market capitalization ceiling, positive earnings, and an opening price near the 10-day moving average. The proposed refinement specifies positive net profits in each of the latest four quarters,…
This article explains how the Fama-French model extends CAPM first with size and value factors, then with profitability and investment factors. It describes double sorts by size and book-to-market, return on equity, or asset growth, and forms RMW from…
This expert-advisor design periodically searches recent price history for RSI or MFI thresholds that produced the greatest simulated monetary profit. It tests buy and sell signals separately across candidate indicator levels, using crossings of oversold and…
The proposed short-term stock screen selects shares with an amplitude above a threshold, an indicated 9:25 price gain below a cap, and an exclusion for the STAR Market. The author presents these filters as a way to combine a volatility condition, an…
The document explains how combinatorially symmetric cross-validation (CSCV) can estimate the probability of backtest overfitting (PBO). It distinguishes training overfitting, where a model performs poorly on held-out data, from backtest overfitting, where a…
BoDi measures the distance between the upper and lower Bollinger Bands and displays that width as a histogram. Because the bands widen as volatility rises and contract as it falls, the author interprets expansion as an active directional impulse and…
The article summarizes Burton Malkiel’s broad investing principles: use index funds, choose capable fund managers, or invest independently after analysis. For individual stock selection, it emphasizes sustained above-market earnings growth, avoiding prices…
This tutorial shows how to use a strategy interface to modify and inspect a running JavaScript trading program. The strategy polls an API for a command, extracts the text after a delimiter, logs the submitted expression, and passes it to JavaScript’s eval…
This post describes a historical stock screen for shares whose codes begin with 60 and whose daily high-to-low range exceeds 1% of the opening price. It restricts observations to 2021 and ranks qualifying rows using a 50-period measure based on the distance…
This introductory article presents ten practical points about machine learning for non-specialists. It emphasizes that machine learning finds patterns in data, so data quality, accurate labels, representative samples, and careful transformations often matter…
The article explains how a long-only smart beta portfolio can combine equity factors to seek steadier returns than relying on one factor. It highlights value, low beta, profitability, investment, momentum, and size, arguing that their different return…
This report evaluates whether changes in listed companies’ employee compensation can help distinguish future stock returns. It compares growth in compensation payable, which represents amounts owed, with growth in compensation already paid, and also…
The document describes a technical stock screen based on daily amplitude, a price threshold, and the close relative to Bollinger Bands. Its stated criteria select stocks with amplitude above 1%, a close below 20, and a close above the middle band but no…