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 brief description presents a single-file collection of averaging methods, including familiar simple and exponential averages alongside smoothed, weighted, triangular, sine-weighted, Hull, and T3 variants. It also lists a linear regression value and…
This indicator description presents Zero Lag Stochastic (ZLS) as a composite of five stochastic oscillators. Each component can use its own %K and %D periods, slowing value, calculation method, price field, and weight; a shared smoothing period applies to…
This stock-screening idea combines a daily price-range threshold, an RSI ceiling, and buying volume greater than selling volume. The range condition is intended to find stocks with movement, while RSI below the stated cutoff avoids the strongest overbought…
The document explains an Ergodic oscillator derived from William Blau’s True Strength Index. It smooths price momentum and absolute momentum through two exponential moving-average stages, forms a normalized momentum measure, then applies additional smoothing…
The document outlines a daily pre-open stock screen for Chinese equities. It selects stocks with MACD above zero and a price below 12 yuan, then ranks candidates by that day’s auction amount and keeps the top five. The rationale is to combine a technical…
The discussion describes a proposed rolling evaluation in which training and testing periods are kept separate. Its example uses successive annual training windows and test windows offset into later years, with each test period’s results concatenated to form…
The post proposes a Chinese stock-selection screen combining price movement, trading activity, and company size. Its stated rules require amplitude above 1, a product involving the previous day’s turnover rate and today’s auction volume relative to the prior…
This brief Chinese-language support note addresses why a filter applied directly in a DataSource may not work as expected. It recommends using a dedicated data-filter module, filtering the resulting dataframe with boolean conditions, or passing a query…
The article questions whether asset pricing and investment can be understood only by predicting returns from company characteristics. It uses the example of rose prices on Valentine’s Day: prices rise when demand is concentrated and sellers are scarce, then…
The document describes a stochastic expansion oscillator as a variation on the conventional stochastic oscillator, with a distinct calculation and display. It identifies three inputs: the %K lookback period, a slowing period, and a noise filter period used…
This document outlines a common set of measures for evaluating trading performance: cumulative and annualized returns, annualized volatility, the Sharpe ratio, and maximum drawdown. It gives mathematical definitions for the return, volatility, and Sharpe…
The document describes an indicator derived from PanPrizMA Sin leverage 72 that calculates a wave’s phase and amplitude. Phase is expressed in degrees across a full cycle, with a complementary antiphase representation; amplitude is measured in price points.…
This literature summary describes applying classification and regression trees (CART) to cross-sectional stock selection. The motivating advantage is that a tree can represent nonlinear relationships and interactions among variables, which linear models or…
The document explains a MetaTrader volume profile indicator that groups trade counts or reported volume by price over chart intervals or a user-selected range. Users can choose range boundaries, data source timeframe, histogram style, and which reference…
This article outlines an approach to improving a random forest stock-selection model through factor screening and parameter experimentation. It first limits the stock universe by excluding firms under special treatment and delisted stocks. Candidate…
This document summarizes experiments with a convolutional neural network (CNN) trading model called Deep Alpha. It reports that a seven-layer version performed better than a two-layer version in a same-period comparison, which the authors attribute to…
A Recursive Moving Average applies repeated smoothing to a price series, with the number of iterations controlling how smooth the result becomes. The description connects this approach to indicators such as double and triple exponential moving averages and…
This Qlib documentation explains a workflow for preparing financial data for quantitative research. Users convert market data into Qlib’s binary format, derive features with its expression engine, apply more complex transformations through data handlers, and…
This document describes a Chinese A-share stock screen combining turnover, market capitalization, profitability, and order-flow measures. It selects stocks with turnover between 3% and 12%, market value below 10 billion yuan, positive average earnings per…
This brief response addresses preprocessing for deep-learning models applied to stock data, including Boolean values and infinite observations. Its direct recommendation for infinities is to remove them. For broader feature preparation, it lists outlier…
This note shows how to construct a simple stock relative-return factor using a precomputed Chinese equity dataset. It identifies one field as each stock’s daily return and another as the CSI 300 index return, then defines relative return as the stock return…
This troubleshooting note addresses errors that arise after adding a rolling-training module to an AI strategy for convertible bonds. It points to a revised notebook and identifies two implementation changes associated with the fix. First, the rolling…
This study applies a factor-timing framework from a 2019 paper to China A-share data. It parameterizes factor weights using external variables and compares two approaches based on where the information enters: a time-series model and a cross-sectional model.…
This short forum post asks how to export a backtest’s daily returns to a CSV file and where to place a suggested calculation in the BigQuant workflow. The proposed approach reads a performance DataFrame, adds one to the cumulative algorithm return, computes…