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 research note examines whether order book data can add stock selection information beyond factors derived from executed trades. It proposes three measures based on changes in net buy orders: the average rate of change, its volatility, and its skewness.…
This brief description outlines a foreign-exchange Expert Advisor that trades from signals generated by the Daniella indicator. A signal is acted on when a colored arrow appears at the close of a bar. The advisor depends on a compiled indicator file, and an…
This overview describes a commodity futures CTA strategy that ranks contracts using a combined score. It considers technical signals—cross-sectional momentum, volatility, open-interest changes, and term structure—alongside macroeconomic measures including…
The environment layer in FinRL-Meta uses cleaned data to create market simulations with a shared, Gym-style interface. Users can build on these environments and compare strategies across a common framework. The document describes account options for margin…
This overview explains how FMZ’s backtests advance through historical data and compares bar-based testing with two tick-oriented modes. In a traditional on-bar setup, each candle creates one decision point, typically using its open, high, low, close, and…
This document explains a script for exporting historical rates and tick data from the symbol shown on the current MetaTrader 5 chart. It addresses a limitation in the platform’s built-in export facility, which may omit available tick history, by allowing…
This document describes an Expert Advisor that trades signals from a universal MACD histogram. It checks for a bar close where the histogram crosses zero, changes direction, or where the signal-line histogram breaks through, with the selected behavior…
This short forum exchange addresses an error in a regression-evaluation workflow that reports a missing label. The response says the problem cannot be diagnosed precisely without details about the task, then recommends first defining what quantity the model…
This Freqtrade strategy builds many Triple Exponential Moving Average (TEMA) series and compares them at intervals controlled by a count and gap parameter. For long entries, it adds conditions when a TEMA is below the preceding TEMA in the selected sequence,…
The document records a user’s question about a trading platform’s order function: sell instructions appear to execute at the following day’s close rather than the day they are issued. The issue affects holding-period exits and closing-price take-profit or…
The document describes a chart overlay for reviewing a symbol’s past trades. It explains the visual legend: blue zones mark take-profit areas, red zones mark stop-loss areas, blue and red arrows indicate buys and sells, and blue ticks or red crosses show…
This strategy places buy limit and sell limit orders in a grid at a configured start hour. Its premise is that after one or more orders activate, price will return to the open of the bar on which the grid was placed. The grid uses a starting distance from…
This Expert Advisor builds signals from two ZigZag indicators operating at different scales. It examines the two legs of the latest peak in the slower, larger ZigZag and counts the smaller ZigZag peaks inside those legs. User parameters compare the peak…
This report reviews quality-focused Smart Beta investing in overseas and Chinese markets. It describes the approach as an index strategy built from company financial data to capture underlying business quality, then compares product development and index…
The document describes a variant of the stochastic oscillator that uses a step chart to filter small changes. A user sets the step size as a percentage of the stochastic value; changes below that threshold are treated as insignificant, which is intended to…
This document describes a stock screen requiring daily amplitude above 1, upward-diverging moving averages, and a price path characterized as arc-shaped. It interprets the shape as a possible rebound after an earlier adjustment. Formula and Python examples…
This indicator profiles five configurable intraday sessions and summarizes their historical behavior for a chosen market. For each session, it records range, volume share, whether it set the day’s high or low, directional closes and continuation, fair value…
This Freqtrade strategy uses a four-hour timeframe and computes an Ichimoku base line and the KST indicator’s difference. Each indicator is normalized across the available dataframe range. Entry and exit thresholds are exposed as tunable numeric values and…
This article explains random forests through CART decision trees, Gini impurity, bootstrap sampling, and aggregation. Trees are trained on resampled data, while their predictions are combined by voting for classification or averaging for regression. It also…
This report compares ways to measure similarity among funds, including cosine similarity, Euclidean distance, and return correlations. It argues that the choice of measure should follow the intended meaning of “similar”: resemblance in holdings structure,…
This report compares ways to measure similarity among funds, including cosine similarity, Euclidean distance, and return correlations. It argues that the choice of measure should follow the intended meaning of “similar”: resemblance in holdings structure,…
This post describes a Chinese equity screen combining positive institutional activity, a threshold for the day's increase in holdings, and a 2021 date filter. It interprets the holding-change measure as a sign of buying interest and positive institutional…
This stock screen combines three conditions: daily trading range above a threshold, return on equity above 15% in each of five years, and a weekly crossover of the five-period moving average above the ten-period average. The rationale pairs a profitability…
This post presents a short-term stock selection filter combining three conditions: prior-day appearance on a market trading activity list, a share price near 18.5 yuan, and an amplitude threshold above 1 percent. It provides example logic for computing…