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.
21,023 documents
This document presents a bounce-strength indicator for assessing buying or selling pressure while price trades inside a range. The proposed measure relates the close’s position within the bar’s high-low spread to its distance from the range boundary: upward…
The Trend Intensity Index (TII), attributed in the document to M. H. Pee, estimates the balance of recent closing prices above and below a longer moving average. For the described 30-day version, calculate a 60-day simple moving average, then compare each of…
The document explains how a simulated trading engine processes each market-data point in three stages: the exchange matches existing orders against the updated market, strategies receive the data and can issue commands, and venues settle eligible commands…
The document describes a ProRealTime indicator that plots how often a chosen variable falls into a set of value bands over a recent sample of candles. The user selects the variable of interest, such as normalized price range or an indicator reading, then…
The document defines the Mass indicator using the high-low range. It first applies an exponential moving average to that range, then smooths the result with a second exponential moving average of the same length. The ratio of the first smoothed range to the…
This forum post presents a DAI query for filtering Chinese A-share records and asks how to save a large result table as an XLS spreadsheet for separate analysis. Its SQL selects dates, instruments, and derived indicators for price-limit events, including…
This essay explains the distinction between zero-sum and negative-sum trading. Before costs, one participant’s gain in a transaction corresponds to another participant’s loss; after commissions, spreads, or other fees, the combined outcome for traders is…
This Chinese-language post describes a stock selection screen using price movement, trading activity, and a short moving-average trend condition. It selects shares whose daily high-low amplitude exceeds a threshold, whose prior turnover adjusted by the ratio…
This document describes a small cross-platform library for comparing floating-point values used in trading code. Its comparison helper applies a chosen precision so that prices or quantities can be checked against thresholds without repeatedly calling a…
The document describes a Fisher Transform oscillator applied to normalized prices. It explains that the transformed series is intended to make price oscillations more distinct, so peaks and troughs in the indicator may help identify possible trend reversal…
This Chinese Q&A discusses basic factor research and strategy refinement. It recommends using a correlation calculation to compare factors and gives rough absolute-correlation bands for weak and strong relationships. It also suggests tracking factor…
This Chinese-language discussion asks why a five-day moving average is missing on a date when the closing price appears to be available and the security was not halted. The response offers a possible explanation: a missing close elsewhere in the five-day…
DUCT is described as an indicator that plots a channel using fractals. Its settings determine the timeframe used to locate fractals and channel control points, the amount of historical data to calculate, and the range of bars searched for fractals. A…
The document reports a failure during a VeighNa CTA backtest when the engine calculates performance statistics. The traceback points to annual downside-risk calculation, where the Empyrical statistics package accesses NumPy’s removed `NINF` attribute. NumPy…
This short quantitative research answer describes a volume-based factor built from two choices: the number of days used to average volume and the number of observations used to measure volatility. Its notation indicates applying a standard deviation to…
This technical indicator displays how many bars have elapsed since the most recent upper and lower fractal formations. It uses two separate lines in a chart window, with one tracking the upper fractal and the other tracking the lower fractal. The description…
The document describes net volume as an indicator intended to represent the balance of buying and selling pressure over a period. A positive reading is presented as indicating that buying pressure exceeds selling pressure, while a negative reading suggests…
This document outlines a Chinese stock screen requiring RSI below 65, year-over-year growth in net profit attributable to parent shareholders above 20% and no more than 100%, and a closing price above the Bollinger middle band but at or below its upper band.…
The document defines the power weighted moving average (PWMA), a moving average in which each observation in the selected period receives a weight based on its position raised to a configurable power. It identifies three inputs: the lookback period, the…
This document extends a cointegration-based spread strategy from pairs to three or more assets. It forms a weighted combination of log prices using a cointegration vector, then derives a spread return from the same weights. Under stated stationarity…
This A-share stock screen combines three conditions: RSI below 65, an outside-to-inside volume ratio above 1.3, and daily position growth above 5%. The stated rationale is to find active stocks that may be developing upward momentum. The document provides…
The document summarizes research on applying conditional generative adversarial networks to calibrate individual systematic trading strategies and combine them into ensembles. The proposed workflow covers training and selecting a cGAN on time-series data,…
The document describes an adaptive bandpass filter intended to isolate market cycles within a selected swing-frequency range while suppressing slower trends and faster noise. It attributes the underlying bandpass-filter concept to John Ehlers and explains…
This article outlines a prototype for automatically collecting after-hours leaderboard data from the Shenzhen Stock Exchange and sending a list of stocks that appear on it. It frames web scraping as a way to reduce manual review of large volumes of exchange…