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 tutorial presents a FinRL workflow for applying deep reinforcement learning to a portfolio of Dow Jones stocks. It formulates trading as a Markov decision process: the agent observes market features and holdings, chooses buy, sell, or hold actions, and…
This FinRL tutorial outlines a deep reinforcement learning workflow for trading a basket of Dow Jones stocks. It models trading as a Markov decision process: the state contains market and indicator data, actions specify share purchases or sales, and rewards…
The report summary presents a Campisi-style performance attribution model for bond funds using net asset value returns, an approach intended for cases where timely holdings data are scarce, especially for private funds. It describes five explanatory factors:…
This user guide describes pysystemtrade as a framework for constructing futures backtests and modifying their components. It covers common tasks such as selecting instruments and date ranges, changing configurations, writing trading rules, inspecting…
This article outlines the strategy layer in an event-driven backtester. A strategy receives market bars from a data handler and emits signal events for a portfolio through an event queue. The design uses an abstract base class with a signal-calculation…
The report describes a stock-selection approach that forecasts the coming information coefficient of seven style factors, then adjusts their weights according to expected effectiveness. Its features include each factor’s historical information coefficient…
This example describes an event-driven stock selection strategy that evaluates companies on the day their financial reports are announced. It selects stocks with year-over-year net profit growth below 1 and ranks candidates by that growth measure. The screen…
This document explains a simple trend-following rule using five-day and ten-day moving averages of closing prices. It buys at the next session's open when the shorter average is above the longer average, then sells at the next open when the shorter average…
This code provides helper measures for selecting groups of four stocks as candidate partners in a vine copula workflow. One traditional method chooses the quadruple with the greatest sum of pairwise correlations. Other methods operate on empirical ranked…
This configuration describes a Qlib workflow for training a LightGBM model and using its forecasts in an enhanced-indexing portfolio strategy. It specifies CSI 300 instruments and Shanghai’s CSI 300 index as the benchmark, with Alpha158 features and…
This event description outlines a defensive quantitative fund of funds approach built around diversification across asset classes. It proposes combining macroeconomic conditions—growth, interest rates, inflation, and credit—with market-level measures such as…
This legacy guide explains how depositing into a Hyperliquid vault gives an investor a proportional share of the vault’s profits or losses. It illustrates the accounting with a deposit that represents one tenth of the vault: when the vault’s value doubles…
This implementation outlines a workflow for building a mean-reversion spread from three or more assets. It fills missing price observations using forward filling or spline interpolation, converts prices to logarithms, and applies the Johansen procedure to…
This report explains a value-versus-growth rotation strategy built around the idea that returns can come from deliberate exposure to style factors as well as broad market risk. It links shifts in style leadership to investor sentiment and risk appetite,…
The article outlines a simple, object-oriented research backtester built around pandas data frames. It separates the workflow into strategy signals, portfolio positions and equity accounting, and performance analysis. Strategies return long, short, or hold…
The report combines risk parity with the Black-Litterman framework to build stock-and-bond and broader multi-asset portfolios. It uses risk-parity weights as the prior allocation, short-term momentum as an investment view, and Black-Litterman to estimate…
The report frames asset allocation as exposure to underlying risk factors rather than simply to asset categories. It distinguishes macro factors, which represent shared risks across asset classes, from investable style factors, which explain risk premia…
This report consolidates tests of 78 factors across eight categories, comparing grouped portfolio returns, statistical independence measures, single-period and cumulative returns, t-statistics, and measures of return variability. It also examines…
This research summary examines whether announcements of institutional investor visits to A-share companies contain information about subsequent stock returns. It describes the events as usually involving specific-company visits, a small number of…
This event description frames index enhancement as tracking a benchmark while seeking excess returns. It outlines a workflow involving single-factor screening, backtest construction, and parameter tuning, and presents a live demonstration that purportedly…
The report examines how index-enhancement funds generate returns above their benchmarks. It separates active returns into industry selection, stock selection within industries, and their interaction, then compares stock selection within index constituents…
This research summary examines industry exposures in existing multifactor stock-selection portfolios, tests how the model performs across sectors, and constructs industry-neutral long and short portfolios. It argues that sector imbalances can weaken…
The document implements Online Portfolio Moving Average Reversion (OLMAR), a portfolio strategy that adjusts asset weights using relative moving-average prices. For each stock, it divides the window’s average price by the current price, then compares each…
This code draft sketches a stock strategy that first obtains a candidate list through valuation filters, then checks a price-based strength condition and MACD. The strength test compares the latest two-day average close with the 30-day average close. Entry…