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 asks whether regression approaches beyond Andersen’s method and Longstaff–Schwartz can estimate exercise decisions for Bermudan claims. It frames the price as an optimization over exercise times and explains that these methods estimate an…
The document compares statistical modeling with stochastic process modeling and calibration in finance. It presents stochastic models as explicit descriptions of how the world behaves, often built to preserve structural relationships such as the absence of…
The document describes an attempt to transform Weibull-distributed observations into standard normal scores using the probability integral transform, with the transformed data intended as neural-network input. The author reports that the resulting histogram…
The document examines an hourly machine-learning signal that predicts the direction of returns measured from an interval’s average price. The key timing issue is that the signal becomes available only after that interval ends, so a backtest that assumes…
The document asks how machine-learning methods such as random forests, support vector regression, gradient boosting, and nearest neighbors compare with GARCH when forecasting volatility beyond a single step. It also asks whether these models use lagged…
The document examines why the Longstaff–Schwartz method assumes American option payoffs are square-integrable, and asks how that assumption relates to finite Monte Carlo samples, regression bases, and path-dependent payoffs. The answer distinguishes the…
The discussion considers ways to search historical stock charts for user-defined or familiar patterns. Suggested approaches include time-series matching, dynamic time warping, classical technical pattern recognition, and unsupervised discovery through…
The document explains particle filtering through a comparison with the unscented Kalman filter. Both methods propagate representative states through a model and update them after receiving a measurement. A UKF uses a small collection of sigma points and…
The document considers whether an R user with existing Monte Carlo backtest scripts should learn quantstrat or continue extending a custom framework. The response recommends treating quantstrat as another available tool, while emphasizing that the choice…
The discussion compares C++, C#, Java, and Python for quantitative finance and argues that language choice depends on the work being done. C++ is described as common for application backbones and computationally demanding tasks such as derivative pricing,…
The document considers whether a point-in-time probability-of-default model can include macroeconomic variables whose effects arrive more than a year later. Its example models next quarter’s default rate using the current default rate and a lagged change in…
The document proposes comparing volatility clustering in return series by fitting hidden Markov models with multiple volatility states. An information criterion such as AIC or BIC can help select the number of states, after which standard HMM estimation…
The discussion distinguishes stationarity from the presence of temporal structure. A stationary Ornstein–Uhlenbeck process can still exhibit mean reversion: when its value moves far from its long-run level, that deviation provides information about its…
The document outlines a supervised learning setup for choosing positions in a single instrument. At each point in time, past prices are summarized as features such as momentum, volatility, and other technical indicators; fundamental information could also be…
The document explains why VIX futures can trade in contango or backwardation and asks whether either shape represents a normal baseline. It relates the curve to expectations about future volatility: distant contracts may reflect the chance of future…
The document explains how to represent a time series with a nonlinear autoregressive mean, a GARCH(1,1) conditional variance, and an ARCH-in-mean effect. The mean equation includes the previous observation, its square, and the conditional variance; the…
The document explores how to build a trading system from traditional chart patterns and machine learning. It raises choices about classification targets, whether patterns should be labels or inputs, and whether a model can identify patterns directly from…
The document considers how to combine coefficient arrays from repeated wavelet decompositions of a time series. The stated aim is to represent the frequency component while leaving residual noise with the smallest standard deviation. The question notes that…
The discussion explains how to include control variables in both the conditional mean and conditional variance of a GARCH model using the rugarch package in R. The example specifies one external regressor for the mean equation and another for the variance…
The document compares two ways to relate an outcome to predictors. A parametric approach specifies a functional form, such as a linear model, and studies whether an estimator recovers its parameters under suitable error assumptions. A second approach avoids…
This document collects recommended books and papers across quantitative finance, including asset pricing, portfolio allocation, option theory, stochastic calculus, derivatives by asset class, risk management, statistics, machine learning, programming,…
The document asks what best practices and regulatory references should guide internal operational risk models for Basel Pillar 2, including models that produce value-at-risk or a similar capital measure. It lists components associated with the loss…
The document asks how to assign returns to five qualitative view levels—very bullish, bullish, neutral, bearish, and very bearish—using standard-deviation thresholds and a time-varying volatility series. It includes an attempted Python procedure that…
The note diagnoses repeated or biased-looking Monte Carlo samples caused by constructing new Mersenne Twister random number generators inside each loop iteration. Its remedy is to create the generator once before the loop and draw successive values from that…