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.
3,481 documents
This forum post describes an attempt to compare a platform’s single-factor analysis module with the open-source Alphalens package. The author uses the CSI 300 universe and the PE_TTM_0 factor, with a 22-day rebalancing interval. They aim to prepare…
This Chinese-language research summary reviews gold’s strong performance through mid-2020 and discusses drivers including heightened pandemic uncertainty and monetary easing. It reports substantial global gold ETF inflows and describes gold’s historically…
This introductory survey organizes multi-agent reinforcement learning (MARL) by the relationships among agents’ objectives: fully cooperative, fully competitive, and mixed cooperation and competition. It explains that cooperative agents can be modeled…
The article examines why sector-themed equity funds attract investors and which industries may suit them, then develops two portfolio design approaches. It links investor demand to the visibility of past winners on financial technology fund platforms and…
This brief platform discussion concerns an error encountered while plotting intraday minute bars for a futures strategy. The response identifies a data-availability issue: fields such as adjustment factors and suffixed close-price columns are not present in…
This market review describes a year of changing styles in Chinese equities: cyclical and growth themes led during the recovery, followed by a shift toward value later in the year. It argues that a high earnings base and flattening growth expectations could…
This industry-selection approach adapts William O’Neil’s CAN SLIM framework by calculating relative strength at the sector level. The report proposes a high relative-strength reading as an early screening signal for industries that may lead the market during…
This study develops an equity factor for company competitiveness across four dimensions: technology, products, internal controls, and capacity to sustain and improve existing advantages. It combines measurable indicators from these areas, then adjusts the…
This equity-factor study compares several ways to measure illiquidity and examines whether less liquid stocks earn higher subsequent returns. Its microstructure measures are quoted spread, effective spread, and price-impact lambda; it also tests Amihud’s…
This article introduces trend following, mean reversion, arbitrage, and high-frequency trading, then argues that trend following is the most accessible of these approaches for individual A-share investors. It describes trend following as entering in the…
The document explains how to construct a DMA feature as the difference between a short-period and a long-period simple moving average. It identifies three inputs: the underlying price or feature, the short window, and the long window. Its worked example uses…
This report examines whether price momentum signals behave differently according to which types of traders account for trading volume. It divides A-share participants into retail, medium, large, and institutional groups based on order size, then compares…
This short data-support exchange explains a discrepancy between stock prices shown by two Chinese market data services. The response attributes the difference to BigQuant providing backward-adjusted prices and notes that providers can use different…
The document raises a portfolio-construction question: how to adjust the weights of several strategies dynamically based on their simulated return curves, instead of assigning fixed weights. It frames the goal as improving the robustness of the combined…
This opinion piece presents trading principles drawn from an interview with a Chinese market participant. It argues for acting on clearly defined chart breaks or confirmed volume-supported moves, using public policy and industry developments to identify…
This page introduces the exploration–exploitation problem in reinforcement learning: a learning agent must decide when to test actions that may reveal useful information and when to choose actions based on what it has learned so far. The lecture is…
BykovTrend presents its signals as colored candlesticks. Green candles indicate rising signals and orange candles indicate falling signals. Candle brightness distinguishes signals aligned with the prevailing trend from those running against it: aligned…
This platform support note explains the intended use of a high-frequency feature extraction module. It takes fields from a specified Level-2, one-minute Chinese stock bar table and applies expressions compatible with pandas and NumPy to calculate…
The discussion explains how to repeatedly reduce a model’s factor list using feature importance. The example starts with 15 factors, removes the three with the lowest importance after a training and backtest run, and repeats the process until five factors…
This document summarizes research on two equity-market questions. The first uses China’s IPO approval process as a natural experiment to examine how a newly approved offering affects existing, substitutable stocks. It reports negative price effects around…
The document explains a Brinson framework for separating portfolio returns into a policy benchmark, active asset allocation or timing, security selection, and their interaction. The benchmark reflects long-term asset-class weights and passive returns.…
This short Q&A shows a fixed-threshold exit rule for stocks and asks how to make the exit dynamic. For each held position, the example retrieves the cost price and latest market price, calculates the percentage change from cost, and closes the position when…
This study asks whether corporate bonds become more vulnerable to price swings when held mainly by open-end funds with illiquid portfolios. It builds a bond-level fragility measure in two stages: first estimating each fund’s portfolio illiquidity from its…
The report compares linear, polynomial, Gaussian, and sigmoid support vector machine classifiers, along with support vector regression, for multi-factor stock selection. Its workflow extracts features and labels, preprocesses inputs, trains and tunes models…