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.
5,018 documents
The document describes an attempt to simulate stock returns, bond returns, and inflation jointly using a Gaussian copula. The author generates correlated normal variables from an estimated covariance matrix, converts them to uniform values, and applies…
The discussion examines how few observed defaults affect probability of default estimates, contrasting simple frequency estimates with Bayesian estimates that depend on a prior. It illustrates the zero-default case with hypothetical credit data, comparing a…
This note compares Advances in Financial Machine Learning by Marcos López de Prado with The Elements of Statistical Learning by Hastie, Tibshirani, and Friedman, focusing on their relevance to quantitative finance. It characterizes AFML as more focused on…
The document examines how the R implementation of multivariable fractional polynomials (MFP) handles continuous predictors in a logistic regression model of fracture outcomes. The author compares the procedure’s cycle-by-cycle output with a textbook…
The document explores why financial researchers often rely on regression and time-series methods despite the availability of flexible machine learning. The responses emphasize that market data are noisy, nonstationary, and limited, while historical patterns…
The document considers how to find recurring motifs in time series. It points to the Matrix Profile as a method designed for motif discovery, while one respondent reports that it did not capture repeating patterns across all scales in their tests. They…
The document asks how to order volume-bar sampling, fractional differentiation, and CUSUM event filtering when preparing financial data for machine learning. The proposed workflow samples trades into volume bars, transforms the resulting series, then applies…
The question asks whether the R package fGarch can fit an EGARCH(1,1) model through its garchFit interface, whose example specifies a conventional GARCH model and allows lag lengths to be changed. The response says that, to the author's knowledge, this is…
The document outlines econometric knowledge useful in quantitative trading, especially for research with financial time series. Its core list includes regression and inference, ARMA and GARCH models, unit-root and cointegration methods, VARs, Kalman filters,…
The document explains that hidden Markov model states for foreign-exchange returns are modeling choices rather than fixed categories such as “hot” or “cold.” One possible setup assigns discrete states according to whether returns are positive or negative,…
The question examines whether an ARCH model’s forecast can be compared directly with a rolling estimate of realized volatility. The example computes realized volatility as the rolling standard deviation of returns, then obtains forecast variance from an ARCH…
The exchange considers whether a recurrent neural network (RNN) trained on one asset’s OHLC history should outperform a multilayer perceptron (MLP) when asked to forecast another asset. The answer argues that naming an architecture alone says little about…
The document surveys ways to estimate multivariate conditional volatility when a Python workflow is desired. It notes that common Python GARCH support may focus on univariate models, and suggests that a basic multivariate model could require custom…
The document asks whether portfolio optimization can be understood as a form of feature selection. It compares assets’ return series, represented as columns in a data matrix, with features in statistical learning. Under that analogy, mean-variance…
The document compares backward-looking returns, which relate a current price to a prior price, with forward-looking returns, which relate a later price to the current price. It frames the distinction as a target-construction issue: historical returns…
The document raises a conceptual question about whether financial time series should be viewed as stochastic processes or as dynamical systems shaped by interacting, changing influences. It uses reinforcement learning as an example: a Markov Decision Process…
The document investigates how a stock-market machine-learning book’s filing index may have been assembled. Its accepted answer proposes that the author scraped or collected quarterly company filing indexes from the SEC’s EDGAR database, then combined them…
The document responds to a student seeking practical guidance for building portfolios with real assets and data. It recommends an applied portfolio-optimization course using R as a starting point, covering time-series work and several portfolio methods:…
The document considers why an intraday linear equity-return model may fit some stocks better than others and whether adding beta, liquidity, or sector returns would make explained-variance statistics more consistent. It outlines factor-based approaches,…
The discussion explains why a machine-learning model with lower mean squared or root mean squared prediction error can still earn less in a backtest. Prediction accuracy evaluates numerical closeness to observed returns, whereas profit depends on how…
The document discusses covariance estimation through the lens of shrinkage and Bayesian methods. Its respondent argues that shrinkage estimators can be understood as Bayesian point estimates with different implicit prior distributions, and recommends…
The document frames the challenge of creating training labels for a machine learning model intended to predict market regime changes. Using the S&P 500 as an example, it describes visually identified bull and bear periods and asks how to classify historical…
The document responds to a question about simulating processes with different Hurst exponent values for research on mean reversion. It identifies fractional Brownian motion as a process parameterized by the Hurst exponent and points to simulation code as a…
The document considers how to simulate future stock prices after backtesting a strategy on historical prices. It critiques a simple return simulation based only on the historical mean and standard deviation, noting that returns form a time series and that…