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.
219 documents
This document presents a C-vine copula method for statistical arbitrage across a cohort of stocks. It transforms daily returns into uniform pseudo-observations using empirical cumulative distributions, selects a C-vine structure by fitting candidates and…
This document is a tabular reference of S&P 500 companies. Its rows pair ticker symbols and company names with GICS sector and sub-industry classifications, headquarters, index entry dates, SEC identifiers, and founding information. It can help researchers…
This Python class appends expanded values to a tabular dataset using polynomial bases or feature products. Its available polynomial families are Chebyshev, Legendre, Laguerre, and ordinary powers; the requested degree controls how many orders are generated.…
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 data-preparation module builds time-series inputs for a deep-learning momentum model. It reads close prices, clips extreme values using an exponentially weighted mean and standard deviation, then derives daily returns and volatility. The target is a…
This document explains how a Kalman filter can update a pair’s hedge ratio over time, avoiding the need to choose a fixed lookback window or a manually selected observation-weighting scheme. It models one asset’s price as a linear function of the other, with…
This note presents a basic stock-selection filter for shares whose codes begin with 60. It requires the daily high-low range to exceed one percent of the prior close and yesterday’s trading value to exceed a stated threshold. The document interprets the…
The code excerpt outlines a Rust live-trading bot builder and event-processing loop. A builder registers instruments with connector and symbol details, tick and lot sizes, and market-depth storage; it can also attach error handlers and order-response hooks…
This technical guide models a positive mean-reverting portfolio as the exponential of an Ornstein–Uhlenbeck process. It fits the model by maximizing average log-likelihood, selecting the portfolio asset ratio that produces the best fit. The model parameters…
The document introduces the Commodity Channel Index (CCI), describing it as a statistical technical indicator that compares price movement with a typical range. It notes that the indicator was first used in futures analysis and later applied to equities. CCI…
This document explains the role of a connector in an algorithmic trading system: it provides a communication point between bots and exchanges, brokers, or market-data providers. A system can manage multiple bots, and each bot can connect to several…
The document recommends plotting live and backtested equity, positions, strategy signals, and order prices together as an initial way to locate discrepancies. If the strategy logic is implemented consistently, it identifies latency and queue modeling as…
This document presents a partial Python implementation of an Alpha101-style factor library. Its functions combine price and volume data using rolling ranks, correlations, covariance, moving averages, standard deviations, price changes, and volume averages.…
The document implements an H-construction approach for analyzing price series, based on a cited study of statistical variability in spreads. It converts a series into Kagi-like turning points or Renko-like threshold steps. The H-inversion statistic counts…
The H-strategy uses Renko or Kagi turning points to study how far a price or spread typically moves before reversing. It defines an H threshold, marks extrema and the later times when a move of that size confirms a turn, then measures the count of reversals,…
These release notes describe Hummingbot 1.14.0, including new centralized and decentralized exchange connectors, documentation changes, and updates to bot orchestration and execution components. The trading-related changes include a KuCoin perpetual…
The document explains an optimal-transport approach to measuring dependence between asset return series. It first transforms observations into empirical copula coordinates using normalized ranks, which removes marginal scales while retaining dependence…
This documentation explains the event record format used by a high-frequency backtesting system and how to validate timestamps in market data. Each record stores an event type, exchange and local timestamps, price, quantity, and optional order or extra…
This overview introduces the Transformer architecture from the paper “Attention Is All You Need.” Earlier sequence-to-sequence systems commonly used recurrent or convolutional networks, often combined with attention. The Transformer instead relies on…
This code presents discretized mutual information (MI) and variation of information (VI) measures for comparing two data series. When the user does not supply a bin count, it estimates one from the observation count and, for the bivariate case, the…
This document describes a Generic Non-Parametric Representation distance for comparing two financial series. It combines a distribution-distance component with a dependence component, with a parameter controlling their relative contribution. The dependence…
This HftBacktest documentation explains how exchange fill rules and queue-position assumptions shape market-data replay backtests. It contrasts a default model that fills orders completely with an alternative that allows partial fills when a trade reaches an…
This document explains a stochastic-control approach to trading two cointegrated assets whose log-price spread is modeled as a stationary process. A mean-reverting spread represents relative mispricing, while a market index and a risk-free asset capture…
This report summary explains a rules-based forecast of constituent changes for four major mainland China equity indexes: the CSI 300, CSI 100, SSE 180, and SSE 50. It uses each index provider’s published construction rules and market and financial data…