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 document motivates adapting forecasting models to changing market conditions: financial data distributions can shift over time, so models trained on earlier periods may lose predictive strength. It compares two approaches, RR and DDG-DA, using linear and…
SwingCyborg is an expert advisor for automating medium- and long-term trades whose trend start and end are chosen by a human. The user provides those turning points, and the software manages the trading continuously. It includes three money-management…
This note describes a Chinese equity screen combining prior-day price range, appearance on the Dragon-Tiger List, and a condition labeled as main-force control. Its stated rationale is that a larger range may signal trading opportunity, list appearance may…
The article introduces a marketplace model in which exchanges and token issuers fund campaigns that reward participants for supplying crypto market liquidity with market-making bots. Rewards are tied to observable order-book conditions: a spread-based…
The strategy defines a range using the high and low of the first four-hour candle of the New York day. Once that candle has formed, it waits for a five-minute close outside the range, followed by a close back inside. A move above the high and then back into…
The report develops stock selection factors from abnormal operating cash flow as proxies for real earnings management. Its premise is that firms can alter real business activity to make reported performance look stronger, potentially harming future…
This expert advisor combines RSI reversal signals with moving average direction. It buys when RSI leaves oversold territory and turns upward while the moving average is rising; the sell setup reverses those conditions. A higher timeframe moving average can…
The script demonstrates a classic allocation strategy that holds a portfolio with a target mix of 60% stocks and 40% bonds. It uses a drift rebalancer: when asset weights move away from their targets by a configured threshold, the strategy sells assets that…
This trading exercise translates a discretionary method into explicit stock-selection rules. A candidate must close above its five-day moving average for three consecutive days and remain within ten percent above its lowest price over the previous twenty…
This note compares portfolio construction approaches, including equal weighting, market-cap weighting, and stratified sampling, with optimization methods such as mean-variance optimization, risk parity, target-risk, ABL, and tracking-error models. It…
This platform guide distinguishes raw financial statement records from processed point-in-time (PIT) data and explains why the distinction matters in quantitative investing. Raw records are organized by instrument, report date, publication date, and change…
This monthly review compares equity factors using information coefficients (IC) and long-short returns. It reports that growth and turnover factors were relatively consistent over the preceding three months, while profitability and reversal results varied;…
This document describes how to configure Lumibot’s CCXT broker for KuCoin. KuCoin is not presented as a globally auto-detected credential route, so the guide uses an explicit broker configuration with the exchange identifier and API key, secret, and…
The document presents a Chinese A-share screening rule combining price movement, large-order net volume, and company size. It selects stocks with an amplitude above 1, large-order net volume above 0.05 for at least three consecutive days, and circulating…
This research summary examines low-volatility investing in overseas markets and mainland Chinese equities. It presents the low-volatility anomaly as a source of long-run risk-adjusted performance, with a behavioral explanation: speculative demand and…
This framework overview explains TqSdk’s component layout and message flow. It describes TqChan as a one-way queue between components and outlines how order messages travel from user code through TqApi and TqAccount to a trading gateway. In the reverse…
This documentation explains how to configure FreqAI within a Freqtrade configuration and strategy. It outlines core settings for training and backtesting periods, model identification, timeframes, correlated pairs, shifted candles, indicator periods, labels,…
This article presents a revised dividend-factor stock selection strategy for Chinese equities and a simpler equal-weight alternative using dividend-themed ETFs. The V2 stock strategy expands the portfolio from 8 to 20 holdings, rebalances monthly instead of…
This example describes two ways to use the relative strength index (RSI) in an equity backtest. Its basic signal rule uses a smoothed moving average to calculate RSI, then takes a short position above 70 and a long position below 30, remaining flat between…
This post presents an A-share selection rule that combines daily price movement, convertible-bond information, and a moving-average trend filter. Its stated criteria include an amplitude above 1, a nonempty name for outstanding convertible bonds, and a…
This Chinese equity screen combines three conditions: MACD above its zero axis, three consecutive down days, and at least two limit-up sessions within the previous 500 days. The accompanying discussion frames positive MACD as a trend filter and prior…
This forum post describes a reproducibility question in a quantitative modeling workflow. The author sets a learning-rate parameter to the same value in the model and in a hyperparameter-search module, and configures the search grid to contain only that…
This post describes a stock-selection screen that combines daily price amplitude above 1, at least one limit-up event during the prior month, and a 2019 dividend ratio above 25%. It frames the conditions as a blend of price movement, recent market attention,…
This proposed EUR/USD strategy on an hourly chart looks for a sharp move through a Bollinger Band followed by a close beyond the opposite band. The long setup also requires the positive directional indicator to exceed the negative one, while the short setup…