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.
62 documents
This article explains how stochastic control models can set dynamic positions in a mean-reverting spread. It outlines two investor preference models: constant relative risk aversion over terminal wealth, and Epstein–Zin recursive utility, which can account…
This document surveys methods for estimating and adjusting covariance matrices used in portfolio risk analysis. It covers the empirical estimator, robust Minimum Covariance Determinant, basic and data-driven shrinkage methods, semi-covariance, exponentially…
This document describes a pairs trading method that uses a two-state Markov regime-switching model to assess whether spread deviations may reflect a persistent change rather than temporary mean reversion. The proposed signal combines the estimated regime and…
This article presents the generic non-parametric representation (GNPR) distance for comparing time series using both distributional and dependence information. The motivation is that correlation or other familiar similarity measures can make series appear…
This document explains history-weighted, or partial sample, regression as a way to make predictions from observations judged relevant to a new input. It defines similarity using negative Mahalanobis distance and informativeness by how far an observation lies…
This article explains why a multi-asset mean-reverting portfolio may be easier to trade when it uses a small number of assets. Sparse baskets can improve interpretability and reduce trading costs; they also avoid the ambiguity that can arise when combining…
The article surveys four online portfolio selection methods that seek to profit from mean reversion: Passive Aggressive Mean Reversion (PAMR), Confidence Weighted Mean Reversion (CWMR), Online Moving Average Reversion (OLMAR), and Robust Median Reversion…
Futures contracts expire at different times, and adjacent contracts can trade at different prices. Joining them without adjustment creates artificial jumps that may be mistaken for signals by a trading model. The note explains how cumulative roll gaps can be…
The article presents a pairs-trading framework that uses Renko- or Kagi-style constructions to identify turning points in a spread. From those points, it derives H-statistics: H-inversion counts directional changes, H-distance summarizes turning-point moves,…
This article compares time, tick, volume, and dollar bars as ways to organize market data for machine learning. Time bars use fixed intervals; tick and volume bars use trade counts or traded quantity; dollar bars use traded value. The proposed rationale for…
The article explains why ordinary bagging can be problematic for financial labels. In event-based datasets, labels may share underlying returns, so observations are not independent. It introduces concurrency to describe overlapping information and uniqueness…
The document introduces Modern Portfolio Theory and explains how asset correlation shapes the risk and return of a portfolio. Expected portfolio return is a weighted sum of asset returns, while portfolio variance also depends on covariances. When assets are…
The document introduces interactive tear sheets for examining candidate trading pairs. It explains why selection requires more than a single cointegration result: Engle–Granger analysis is sensitive to which asset is treated as dependent, while Johansen…
This release announcement describes changes to MLFinLab, a toolkit for developing machine learning based trading systems. Bar generation now returns timestamps as a DataFrame index, aligning its output with downstream functions and avoiding manual index…