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 MetaTrader 5 utility analyzes closed trading history to show how reported performance was formed. Alongside net and gross results, profit factor, win rate, and average and median trade, it examines reliance on the best trades, sensitivity to removing…
This strategy takes a long position in an S&P 500 proxy on the fifteenth calendar day of each month, shifting the date earlier when it falls on a weekend. The position is intended to last through that trading day, with liquidation on the following scheduled…
This trading-system description uses the XFatlXSatlCloud indicator to signal a change in trend direction. A trade signal is generated at bar close when the indicator cloud changes color. The document concerns an Expert Advisor setup, requiring the compiled…
This Freqtrade strategy generates long-entry and exit signals by comparing configurable dataframe columns. Entry occurs when a selected fast series, optionally shifted by a chosen number of bars, crosses above a selected slow series multiplied by a vertical…
The document outlines the historical datasets available for Hyperliquid and how advanced users can retrieve them from public cloud storage. Market data includes level-two order book snapshots, while asset context files are stored separately. The archive is…
This summary reports tests of 19 factors derived from sell-side analyst consensus data, covering forecast financial measures and analyst attention. For financial factors, it compares raw values with industry-relative rankings and quarter-over-quarter…
This Chinese-language post describes an equity screening rule that combines a daily price range threshold, a named morning star candlestick pattern, and at least one large single-day gain within the recent 25 trading days. It frames the screen as a way to…
The post outlines a Chinese equity screening idea combining daily price range, a low share-price threshold, and recognition of a Morning Star candlestick pattern. Its example implementations also apply a Bollinger-band closing-price condition, then show…
This post proposes screening equities for daily amplitude above 1%, observations dated in 2021, and a price-change condition below 6%. It describes high amplitude as a sign of larger price swings and treats a modest historical gain as a possible entry…
This Chinese-language post outlines a screen for mainland Chinese equities in the metaverse theme. It selects stocks reported on the previous day’s trading list and requires the 30-day moving average to be rising versus its prior value. The post also…
This post outlines a daily long-short strategy for the S&P 500 using a two-period RSI. It opens a long position when RSI falls below a parameter and a short position when RSI exceeds a mirrored threshold. Long exits require RSI to recover to a separate…
This document outlines a machine learning workflow built around feature engineering, model training, and model combination. Feature engineering includes creating, extracting, and selecting variables. It contrasts linear models, which capture linear…
The article examines a z-score indicator whose automatic lookback selection is described as machine learning. It selects among ten fixed windows using price-trend fit, then trades z-score extremes as mean-reversion signals. The audit identifies problems in…
This brief project note reports a modification to an existing quantitative strategy: market capitalization was added to the scoring process and return on equity was removed. The author says the backtest improved after the change, but provides no performance…
The document describes an Expert Advisor that uses recent Bollinger Band Width values in a neural network method. Band Width is presented as the percentage spread between the upper and lower Bollinger Bands, normalized by the middle band in the…
The document presents a calculation for charting the annualized return of a fund over time. It retrieves daily closing prices for a specified fund and date range, then compares each later close with the first close to calculate cumulative return. It…
This BigQuant example attempts to close a stock position after its price falls more than 10% below cost, then exclude that instrument from target holdings until a cooldown expires. It tracks expiration dates in a pool, removes expired entries each day,…
The document describes a trading system that combines the Alligator indicator with Fractals. Its basic entry idea is to detect a particular grouping of Alligator lines and confirm the signal with a fractal. The system also offers configurable indicator exits…
The post describes a Chinese equity screening rule combining daily price movement, circulating market capitalization, and a rounded or arc-shaped chart pattern. Its stated conditions are amplitude above 1, circulating market value above 10 billion yuan, and…
This introduction to crypto pairs trading argues that correlation alone does not establish a durable relationship between two assets. A pair may move together because of shared market forces, yet its price spread can continue drifting. Cointegration offers a…
This note describes a Chinese A-share screen requiring an RSI below 65 and seven consecutive sessions in which the close is no higher than the open, followed by a ranking based on a stock popularity measure. It outlines the intended use of technical…
The author describes a monthly stock-selection approach that buys the ten lowest price-to-book stocks in the CSI 300 universe, then combines it with a market-timing signal based on a hidden Markov model. The starting idea is a single-factor value strategy,…
The report compares ordinary K-fold cross-validation with time-series cross-validation for tuning machine-learning models on sequential data. Random folds can place later observations in training while earlier ones are used for validation, creating leakage…
This configuration describes a Qlib experiment for CSI 300 equities using a LightGBM regression model. The dataset combines daily features with intraday one-minute data resampled to a daily frequency through a custom handler. The configured label frequency…