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 strategy ranks constituents of the CSI 300 by their trailing 12-month dividend yield and selects the ten highest-ranked stocks. It updates the target list on the first trading day of each month, sells names that leave the selection, and buys newly…
The report extends a single-asset drawdown adjustment approach to portfolios of strategies. It describes selecting the control parameter gamma directly from the single-dimension result, avoiding Monte Carlo optimization for the multidimensional case. The…
The article explains recurrent neural networks and LSTM gates, then outlines an A-share stock-selection example that predicts five-day returns from sequences of seven factors. It describes missing-value handling, outlier filtering, standardization, and…
This article introduces a custom-coded stock selection strategy built around five rules, mainly seeking stocks that have fallen sharply and begun to stabilize, with some rules intended to capture possible strong buying. It contrasts rule-based strategies,…
The article explains how risk-constrained Kelly sizing modifies the classic Kelly approach, which chooses a capital fraction to maximize long-run log growth but can produce large drawdowns. The constrained version adds a limit on the probability that wealth…
This document describes a trading system that makes decisions when the lines of a Leading indicator cross. It identifies the crossover as the system’s entry signal, but does not explain the indicator’s calculation, specify how the two crossover directions…
This overview introduces regression as a way to model relationships between predictors and outcomes, then compares seven techniques: linear, logistic, polynomial, stepwise, ridge, lasso, and ElasticNet regression. It describes common use cases and…
The document outlines a framework for allocating across asset classes according to macroeconomic trends. It first identifies whether individual indicators are rising or falling, using historical averages or filtering, then compares asset returns over the…
This brief notice outlines a composite scoring framework for strategies listed in a marketplace. It assigns weights to three broad categories: performance, risk, and risk-adjusted performance. The listed performance measures are cumulative return and success…
This short platform discussion addresses whether a backtest can sell existing holdings at the open and use the proceeds to buy other stocks at that same open. The questioner reports that setting both order prices to the open did not make sale proceeds…
This study compares time-series and cross-sectional standardization of price and volume factors in an A-share stock-selection setting. It uses a dataset spanning 2011–2022 with 98 daily factors and a five-day return label, and compares LightGBM and deep…
This documentation explains how to define model inputs and prediction targets in FreqAI strategies. It distinguishes base features, such as price indicators, volume, and time variables, from configuration-driven expansions across periods, timeframes, shifted…
This ProRealTime version 11 indicator builds a historical distribution of price changes by comparing each close with the price from a configurable number of bars earlier. It separates gains and losses, rounds percentage moves into buckets of 0.1 percentage…
This case study describes a hedge fund seeking to add digital asset strategies and the data infrastructure needed to research and trade them. Its requirements included real-time and historical market data, high-volume feeds for algorithm development and…
This configuration describes a Qlib workflow that trains an XGBoost model on China A-share data for the CSI 300 universe, using Alpha360 features. Its label is a forward close-to-close return over the next interval. Training and validation use earlier…
This document explains how to run a ZigZag extrema calculation on a source buffer, illustrated with an RSI buffer in an indicator. The function writes detected turning points to an output buffer and accepts a search mode for highs, lows, or both. It can also…
The document presents a framework for testing stock-selection factors. It evaluates each trading day’s stock factor values against returns over subsequent holding periods, aiming to show how predictive information changes as the holding period grows. Daily…
A short VeighNa community discussion asks whether performance statistics are calculated from total account capital or from the capital associated with the traded instrument. The poster observes that when starting capital is large relative to fixed order…
This Chinese-language strategy note proposes screening equities for a specified turnover range, a 2021 listing year, and a weekly five-period moving average crossing above the ten-period average. It presents the crossover as a technical momentum filter and…
This Chinese-corn-futures example combines the Chande Momentum Oscillator (CMO) with a short moving average to generate long and short entries. Signals include reversals from overbought or oversold levels, CMO crossings of its signal line, and zero-line…
This strategy describes an hourly DAX breakout approach. It enters long when price crosses above a prior Donchian channel high, with the MACD histogram rising and nonnegative, RSI above a threshold, and price above a moving average. A stop is placed at the…
This example demonstrates borrow conflicts that can occur when Python actor callbacks reenter signal operations in a backtesting engine. A subscribing actor subscribes during startup and, in the signal scenario, publishes a nested signal from its callback…
PhantomFlow identifies trend from market structure rather than smoothing price and reading its slope. It detects five-bar fractal highs and lows, which are confirmed two bars after the swing, then applies a zigzag filter so accepted pivots alternate between…
This stock screen combines price movement, trading activity, and historical performance. It selects shares with a stated amplitude above one, prior-day actual turnover between 3% and 28%, and a 2021 gain above the market average. The article’s rationale is…