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 introductory tutorial explains machine learning concepts before focusing on linear regression. It defines features and labels, distinguishes supervised from unsupervised learning, and outlines classification, regression, clustering, and parameterized…
This Chinese-language post outlines a stock-selection screen based on three conditions: a positive but limited ten-day price gain, net large-order inflow during the afternoon, and at least five moving averages converging. It describes selecting a stock…
This expert advisor makes a directional decision when a new bar appears by comparing the tick volumes of the two most recent bars. If the newer bar has greater volume, it closes sell positions and opens a buy; if volume is lower, it closes buys and opens a…
The document explains Bollinger Bands as a moving average surrounded by upper and lower envelopes set using a multiple of recent closing-price standard deviation. The center line is an N-day average; the width changes with measured price dispersion. It…
This weekly report presents diagnostics for the ten highest-ranked anonymous factors in a Chinese quantitative research competition. It describes average style exposures estimated with daily multivariate regressions and average industry-level Rank ICs, the…
StockRanker is a tree-based learning-to-rank tool for ordering stocks from supplied features. Its listwise approach optimizes the quality of a stock list’s ranking, and the guide connects the method to the RankNet, LambdaRank, and LambdaMART line of work.…
The document addresses a backtest or trading-log puzzle: a buy order is calculated by rounding available cash down to a whole multiple of 100 shares, yet the recorded final purchase quantity is not a multiple of 100. The explanation given is that the system…
The document explains two differences between backtests and simulated or live trading in a notebook-based platform. In the described live modes, the initialization function runs during each trading period, whereas a backtest runs it once. It also says…
This update log describes changes to a Chinese equity screening expression. The initial filters included a trailing price-to-earnings ceiling and a minimum listing age. The author then replaces a fixed floating-market-capitalization threshold with a…
This MetaTrader 5 indicator adapts an earlier MetaTrader 4 tool that searches price history for candle sequences resembling a selected recent pattern. It uses dynamic time warping (DTW) to identify candidate matches and presents possible future price paths…
This document lays out an ordered process for adding a strategy to a live trading system or replacing an existing one. It covers preparing instrument data, confirming a working backtest, configuring strategy and control files, implementing custom backtest,…
The article uses binomial price paths to explain how adding to positions changes the distribution of trade outcomes. In a fair random walk with zero expected return, it shows that both adding in the direction of the move and adding against it leave expected…
This guidance explains why AI-generated strategy logic, code, reviews, backtests, and parameter recommendations should be treated as research aids rather than final trading decisions. Outputs may contain errors or omissions, vary across models or settings,…
This brief VeighNa community exchange answers how to inspect variables during a backtest. It suggests printing values or writing them to the system log, giving two straightforward ways to observe runtime state while a strategy runs. For a backtest started…
FreqAI is presented as an open source framework for training machine learning models to forecast market targets from user defined indicators. Users supply features and future looking labels; the framework trains a model for each listed trading pair and…
The article introduces algorithmic trading as a way to automate rule based market analysis and order execution. It describes potential uses including applying technical indicators, high frequency trading, arbitrage between markets, scalping small price…
The post describes a user’s issue with a simulated cryptocurrency spot backtest: its performance overview appears to classify buys as idle periods and does not record sells. The sample strategy uses a moving-average crossover, tracks position state with a…
This article outlines how institutional crypto teams can assess whether their market and on-chain data coverage supports trading, research, compliance, and risk work. It links potential data gaps to symptoms such as missed arbitrage opportunities, difficulty…
This post outlines an equity screening rule using a daily high-low range threshold, sustained return on equity, and a bounded turnover ratio. Its rationale is to combine a minimum level of price movement with evidence of profitability and a middle range of…
This post proposes a Chinese equity screening rule that combines daily price range, a recent three-session limit-up streak, and simultaneous bullish crossovers among short and longer moving averages. It presents the rule as a way to pair evidence of strong…
This function creates a synthetic binary classification dataset for studying feature importance and redundancy in an asset-management machine-learning context. It first generates informative and noise features, then constructs additional redundant features…
This overview catalogs FMZ platform tools for operating and developing trading bots. It describes bulk starting and stopping, log alerts, persistent bot identifiers, parameter import and export, strategy grouping, editor options, and remote editing. For…
The post describes a stock screen combining a 14-period RSI below 65, three consecutive candle conditions, and a weekly price crossover involving a 30-week moving average. It presents the setup as a way to find stocks that may rebound after weakness, and…
This review describes a model in which equity factors move through normal conditions, periods of value weakness, and subsequent reversals. The value drawdown may reflect a rally in growth stocks or a collapse in value stocks; reversals then differ according…