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.
5,922 documents
The article summarizes research on whether information spreads through mutual fund families and how that relates to fund performance. It describes a diffusion measure based on the order in which affiliated funds buy a stock after it first appears in a family…
This document describes a daily Chinese equity screen that ranks stocks by free-float market capitalization, turnover rate, and closing price, combines the rankings into a composite score, then selects the lowest-ranked names. It filters out…
This research digest summarizes three studies. The first compares earnings announcement returns (EAR), which capture market reactions to unexpected information in company results, with standardized unexpected earnings (SUE). It reports annualized long-short…
This report examines two style-allocation ideas for Chinese equities: rotating between value and growth using value spread, and reducing market exposure to systematic risk through low-beta investing. It argues that value spread tends to revert toward its…
This essay uses hypothetical investment examples to explain how returns compound asymmetrically: a loss requires a larger percentage gain to recover, and alternating gains and losses can produce a modest long-run result despite large individual moves. It…
The document explains an account-state object used to represent a snapshot of balances and margin information. Such snapshots can arrive from venue updates through an execution client or be calculated by a portfolio after a position change when calculation…
The document outlines a way to weight stock positions using a momentum signal and an estimate of volatility. It describes two alternatives: an exponentially weighted moving average based on squared returns, and a Yang–Zhang variation of Garman–Klass…
This introductory article explains the basic objects and notation of linear algebra used in machine learning and deep learning. It defines scalars, vectors, matrices, and higher-order tensors, with examples such as feature vectors, neural-network weights,…
This research summary reviews six studies on industry rotation and index enhancement in China. Its multi-factor framework considers expected and historical fundamentals, price and volume information, market leadership, and system-risk concentration. It…
This introductory article explains how rapid price rises fueled by investor enthusiasm can develop into cryptocurrency bubbles and sharp declines. It identifies new investor inflows, fear of missing out, media attention, and imitation of other buyers as…
This report adapts Peter Lynch’s six company categories to China’s A-share market: slow-growing, stable-growing, fast-growing, cyclical, turnaround, and asset-rich companies. Its selection framework first screens for a positive price-to-earnings ratio and a…
This configuration defines a Qlib workflow that trains a PyTorch feedforward neural network on Alpha360 features for CSI 500 stocks. It uses a forward close-price return label, robust feature normalization, missing-value filling, and cross-sectional rank…
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…
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 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…
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…
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 document introduces factor research as a quantitative way to study characteristics that may help explain asset returns and risk. It names market capitalization, quality, momentum, low volatility, and yield as examples, and describes using statistical or…
This report explains how the MSCI China A50 Connect Index selects and weights large-cap Chinese equities. It chooses stocks across GICS industries, first taking leading free-float market-cap names by industry and then filling to 50 constituents; weights are…
This introductory guide lays out five areas for developing quantitative trading capability: mathematics and statistics, programming, financial knowledge, strategy research, and practical testing. It highlights time-series and cross-sectional econometrics,…
The document addresses how to combine multiple stock-selection factors when they have different preferred directions. Its example asks how to score one factor from low to high and another from high to low, then rank securities by the average of their scores.…
This guide recommends choosing community trading strategies according to an investor's capital, return expectations, and tolerance for drawdown. It distinguishes a conservative profile, associated with diversified holdings, a smoother equity curve, a longer…
The report describes an asset allocation approach that measures cycle states and uses machine learning to estimate the probability that assets will outperform one another. It reviews macroeconomic timing frameworks, then describes cycle factors derived from…
This configuration specifies a Qlib workflow for training the HIST model on Alpha360 features for CSI 300 stocks and evaluating its signals in a portfolio backtest. The data handler normalizes features, fills missing feature values, drops rows without…