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
The proposed Chinese stock screen filters for daily turnover between 3% and 12%, excludes Beijing-listed A shares, and selects stocks described as being controlled by major holders on the previous day. The article discusses this holder-control signal as a…
The document explains an approach to mutual-fund evaluation that adds an active peer benchmark (APB) to standard asset-pricing models. The APB is formed from the average returns of funds in a comparable style group, providing a proxy for shared strategies…
The document summarizes a macro-factor framework for strategic asset allocation, illustrated with portfolios for endowments, life insurers, and public pension plans. It lays out four stages: derive macro factors from principal components of monthly returns…
The document defines the Historical Volatility Ratio (HVR) as the ratio of two standard deviations calculated over configurable periods. Each standard deviation is applied to the log price change, defined as the logarithm of the current applied price divided…
This forum post reports a possible data-handling issue in a bar generator that aggregates ticks into one-minute bars. When the first tick arrives just after 9:30, its last traded price is used to initialize the bar’s open, high, low, and close. The post says…
The Absolute Strength indicator measures the relative strength of buyers and sellers in points, presenting it as an alternative to RSI, Stochastic, and DMI. The description says point-based values can show finer-grained changes in strength than…
This module implements the bivariate Nelsen 13 copula, a tool for modeling dependence between two uniform variables. It provides the copula cumulative distribution and density, a conditional distribution, random pair generation, and a parameter estimator…
This research note introduces time-series stationarity as preparation for studying pairs trading in cryptocurrency futures. It explains weak stationarity through stable mean and variance and covariance that depends on the time gap rather than the observation…
This document describes a utility library for working with floating-point numbers in trading software. Its functions cover approximate comparisons, rounding to decimal places or increments, normalization and range mapping, clamping, safe division, binary…
This overview presents BERT, a language representation model pretrained with context from both the left and right sides of text. It contrasts that design with earlier unidirectional language models, which restrict each token to information from preceding…
This note describes a trend indicator based on the correlation between a security’s price history and an idealized straight-line trend. The underlying approach, attributed to John Ehlers, is intended to identify the beginning of trends and recognize when…
This article presents quantitative trading as a tool increasingly used by professional and institutional market participants. It argues that rule-based models can complement fundamental and technical analysis by screening many stocks against user-defined…
This proposed equity screen combines three technical conditions: daily amplitude above 1, a rising price floor, and a ratio based on the previous day’s turnover multiplied by today’s auction volume relative to the previous day’s volume. The ratio must fall…
This Chinese-language research summary describes constructing stock-selection factors from tick trade records. Rather than relying only on the buy-sell aggressor flag, it uses the associated buy-order and sell-order identifiers to reconstruct orders from…
This notebook explores relationships between Colombian peso exchange-rate data and crude oil blends, currencies, and gold. It ranks simple ordinary least squares regressions by R-squared, compares the Vasconia crude relationship with the peso before and…
The HighestLowestRange indicator is presented as a histogram that marks entry into overbought and oversold zones with color changes. Its first bar in a newly entered zone receives a brighter color when the move follows an opposite trend, drawing attention to…
This Chinese equity screen combines trading activity, reported profit growth, volume, and opening price. It selects stocks with turnover between 3% and 12%, parent-company net profit growth above 20% and no more than 100%, current volume above 10,000 lots,…
MultiVote OBV modifies standard On Balance Volume by using changes in the high, low, and close rather than relying on closing-price direction alone. For each bar, it assigns a positive vote when a price component rises from its previous value and a negative…
This brief exchange explains how to interpret a feature-importance chart when one or more short-horizon factors receive a score of zero. The response says the chart ranks factors within the particular set used to train a particular model. A low-ranked factor…
This document describes a data preparation workflow for daily risk estimates on China A-shares. For each date, it selects the CSI 300 constituents, gathers a rolling window of closing prices, calculates returns, and clips extreme returns at the…
Candle_Ratio is an indicator that compares bullish and bearish candlestick movement over a chosen lookback period. For bullish candles, it records the close-minus-open amount as upward movement and assigns zero downward movement. For bearish candles, it…
This document describes an indicator that adds a trend-strength display to the Gaus_MA moving average. It uses the standard deviation of the indicator and two user-set thresholds to classify the reading. When standard deviation lies between the thresholds, a…
The document gives a brief historical and conceptual overview of reinforcement learning. It starts with Markov decision processes and dynamic programming, then describes temporal-difference learning and Monte Carlo methods as ways to estimate value without…
The document describes a stock screen for companies whose codes begin with 60, with turnover between 3% and 12%, market capitalization below 10 billion yuan, and no losses in each of the prior three years. It frames these filters as a way to combine…