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
This strategy ranks commodity futures by an empirical measure of return asymmetry. At each monthly rebalance, it counts daily returns over the previous 260 trading days that exceed the mean by more than two standard deviations, then subtracts the count below…
This guide explains how long-short equity portfolios combine stocks expected to rise with stocks expected to fall. It discusses possible return sources, including security selection, market timing, and exposure to factors such as value, growth, momentum, and…
The document compares two ways to estimate a bond fund’s duration. A holdings-based estimate builds a proxy portfolio from the largest positions disclosed in periodic reports and calculates its weighted duration. A NAV-based estimate regresses the fund’s net…
The document summarizes a slack-based manager efficiency index (SMEI) for evaluating mutual fund managers as a whole rather than judging each fund in isolation. The data envelopment analysis framework treats risk, portfolio turnover, and a combined…
This document outlines a China A-share stock selection strategy using market, size, and book-to-market factors. It starts with constituents of a benchmark index, filters out stocks that are suspended or marked as special treatment, and uses a historical…
This assignment outlines a style-rotation process built from three stock strategies: small-capitalization, below-book-value, and oversold-rebound approaches. After backtesting each strategy, it proposes saving the return and position series, then comparing…
This document explains industry rotation as a strategy that shifts equity exposure toward sectors showing stronger performance and may reduce equity holdings when conditions look unfavorable. It discusses possible drivers, including economic cycles, policy,…
This report examines how portfolio optimization settings affect a multi-factor equity strategy. It considers the interaction of active-weight limits with the lambda coefficient, the share of non-index constituents, industry and size exposures, turnover…
The post describes a China A-share stock screen for metaverse-related companies listed on the Shenzhen main board. Its stated filters combine a minimum circulating market capitalization with positive price-to-earnings and price-to-book ratios capped at…
This implementation describes a pairs-trading method based on modeling the log price relationship between two stocks as an Ornstein–Uhlenbeck process. It constructs the spread as the difference between the stocks’ log prices, fills missing observations…
This document summarizes lifecycle funds, which adjust their asset allocation as a target date approaches. The described approach gradually reduces portfolio risk over time, with the aim of matching investment risk to the holder’s changing capacity for risk…
This short exchange explains how to interpret a long-short portfolio return in factor analysis and how to select a chosen group from a factor-ranked universe. It describes the long-short spread as the return from buying the lowest-ranked group and selling…
The post outlines a proposed equity workflow that links market breadth, sector selection, stock screening, and portfolio exposure. It would compare broad-market indices with industry or concept indices using beta, select sectors with higher beta, then choose…
The article describes inferring a fund’s constituents and weights from its total value time series. It compares Extended Linear Cloning (ELC), which uses regression coefficients as asset importance scores and retains the highest scoring candidates, with…
This code utility builds pairwise dependence matrices from columns in a feature DataFrame. It supports information-based measures, distance correlation, rank correlation, GPR and GNPR distances, and optimal-transport dependence. Parameters let users…
This overview explains how AI quantitative investment platforms combine machine learning, deep learning, and natural language processing with financial data analysis. It describes a general workflow in which platforms collect and process market and financial…
The article describes a framework for identifying crowded trading and distinguishing a bubble’s buildup from its unwinding. It estimates asset clustering from covariance and principal components, using the concentration of sector returns in important…
This module describes a trading rule built around a pre-estimated multivariate cointegration vector. It calculates the weighted sum of log prices, differences that series across recent observations, and uses the sign of the summed changes to set trade…
This study proposes a passive index that selects between the A-share and H-share of the same company according to which listing is cheaper after currency adjustment. It examines price gaps across dual-listed Chinese companies, using panel regression to…
This research note describes extending mean-variance portfolio allocation with higher-moment information, especially co-skewness and co-kurtosis among assets. Mean and covariance alone can miss differences in joint tail behavior: portfolios with similar…
This report summary lays out a framework for analyzing government interest-rate bonds through supply and demand, return decomposition, duration timing, and portfolio allocation. It attributes supply partly to fiscal conditions and economic pressure, and…
This research summary explains a dynamic firm model in which companies can expand or shrink their operating scale, but face adjustment costs. Expansion and contraction options alter exposure to productivity shocks: the ability to reduce scale can buffer…
This essay argues that quantitative, discretionary, short-term, long-term, active, and passive approaches coexist within a market ecosystem. It warns that historical models can fail in unusual conditions, crowded strategies may lose their edge, and…
This A-share stock-selection approach combines an earnings-surprise proxy with the price response to earnings news. Instead of comparing reported profit with a numerical consensus estimate, it analyzes analyst research-report titles for wording that signals…