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
The strategy first switches between the CSI 300 and the ChiNext index based on the changing gap between their performance curves. It describes a widening gap as a signal to buy the stronger index and a narrowing gap as a signal to buy the weaker one, using a…
This strategy maintains an equal-weight portfolio of Bitcoin and Ethereum using daily checks and monthly rebalancing. At each rebalance, it estimates target quantities from current portfolio value, divides value evenly between the two assets, and submits buy…
The document outlines a China A-share screening rule that combines daily price range, a newly formed KDJ crossover, and a proxy for institutional buying. Stocks enter a candidate pool when the high-to-low range exceeds 1%, the J line crosses above D, and…
The document reviews research on whether crowded mutual fund ownership predicts subsequent stock returns. It defines stock-level crowding as active mutual fund ownership divided by average share turnover, using lagged holdings and turnover data. Stocks are…
This document describes the first BigQuant strategy development interest group, scheduled for May through July 2024. Its purpose is to help learners build strategy templates and demonstrations, reproduce common methods, and improve their implementation…
This document describes an Expert Advisor that trades when the Go indicator changes color at a bar’s close. It opens a trade when the bars switch from red to green or green to red. The system relies on a compiled indicator file, and the article notes that a…
This proposed stock screen first restricts the universe to shares classified in the metaverse industry. It then looks for an opening price around the ten-day moving average and ranks qualifying shares by the day's auction amount, selecting the top five. The…
This stock selection proposal combines an RSI threshold below 65 with a positive product of price change and a large-order net-flow measure, and requires the daily high-low range to exceed one percent of the previous close. It presents these filters as a way…
This ProRealTime strategy targets an overnight rise in the FTSE MIB by entering long at 5:30 p.m. and exiting at 9:30 a.m. It adds two entry conditions: the 9:30 a.m. open must be below the previous daily close, and the 5:30 p.m. price must be at or below…
This configuration describes a Chinese equity forecasting experiment using Qlib’s Alpha158 features and an LSTM model. It selects 20 features, applies robust feature normalization and missing-value filling, and ranks labels cross-sectionally. The prediction…
This research summary tests whether industry returns left after removing common style exposures can help rank sectors for rotation. It reports a strong difference between markets: an alpha based on the Fama–French three-factor model works for US industries,…
The document describes an automated trading system based on the BlauTStochI oscillator. It identifies two alternative entry approaches: trade when the oscillator histogram changes direction, or act when it crosses the zero line. An input setting selects the…
This research describes a systematic trend-following strategy for commodity futures. It first uses MACD crossings, filtered by cumulative distance relative to ATR, to classify rising and falling trends. It then revises those classifications when price…
This document describes a long-or-short crossover strategy that searches pairings among 70 moving average and filter types for a chosen market and timeframe. The selected series use a common period, and an additional exponential smoothing step is intended to…
This post presents a Chinese stock-selection screen that filters for turnover between 3% and 12%, requires at least one limit-up event in the previous 25 days, and ranks candidates by the day’s opening-auction amount, selecting five. The accompanying…
This recap presents four trading scripts developed by participants in an educational program. The examples include a portfolio that rebalances equally across assets, spot–perpetual arbitrage that opens positions when the price spread exceeds a threshold and…
The document defines a price-based momentum factor as the 44-day average closing price divided by the current close, minus one. It compares two ways to analyze the factor: using a platform expression engine and supplying custom data. The stated test covers…
The document describes a stock screen combining turnover, industry classification, and the day’s lowest price move. It selects beverage and alcohol related companies with turnover between 3% and 12%, whose intraday low is between 4% and 5% below the previous…
This article describes a Chinese equity screening idea that combines a relative strength index threshold, membership in the beverage and alcohol import-export industry, and an upward-moving 30-day average. It then proposes a stricter version: require the…
The article uses gambling and trading stories to explain why survival and capital preservation come before pursuing returns. Roulette illustrates how a small house edge compounds with repeated play, while blackjack card counting shows how a positive edge can…
This stock screen combines three technical conditions: amplitude greater than 1, a positive weekly MACD histogram, and MACD above zero. The article frames the range and weekly histogram conditions as signs of stronger price action, while the zero-line…
This article presents ten general practices for developing a disciplined crypto trading process. It recommends starting with simulated trading, moving to small live positions, setting measurable progression goals, and backtesting strategies before…
This Chinese-language forum post asks how to interpret floating profit and loss, period profit and loss, and capital utilization in a backtest chart. It supplies a dated sample table showing those fields alongside trade volume and long and short positions.…
This Expert Advisor follows the HullTrend indicator and generates a signal when a completed bar changes the indicator cloud’s color. The description presents this as a trend-following rule and identifies USDJPY on a four-hour chart as the setting used for a…