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 post asks how to fit a skew-normal or skew-t copula to data while allowing the correlation matrix to change over time. It highlights an implementation challenge: the author has not found a suitable approach in R or Matlab and is open to multiple…
The document considers how team based quantitative challenges can build skills in programming, residual analysis, portfolio management, allocation, statistics, and decision making under changing objectives. The original prompt seeks research exercises using…
The document asks how to estimate one response variable from many explanatory variables, such as a set of daily economic indicators. It contrasts a brute force or genetic optimization approach with ordinary least squares and notes that linear algebra…
The discussion addresses how repeated runs of stochastic optimization can produce changing stock rankings, even after the model has converged. Suggested approaches include running multiple instances with different random seeds and aggregating their…
The document collects suggestions for studying the mathematical foundations and newer approaches to Value at Risk (VaR). It mentions combining multiple VaR methods with artificial neural networks, with the network balancing weaknesses among component…
The document introduces reject inference in retail credit scoring: estimating default behavior for applicants who were denied credit and therefore lack observed repayment outcomes. It notes that parceling is one established approach and points to a bivariate…
The document asks how to convert daily return forecasts and their forecast variances from a Kalman-filtered stock-market model into an investment allocation. It defines forecast return, realized return, and forecast error, and assumes realized return is…
The document considers regression when a dependent variable responds differently to increases and decreases in a predictor. It notes that a dummy variable and interaction with the predictor may not resolve the issue, then raises a separate concern from the…
The answer shows how to combine an autoregressive moving-average model for the conditional mean with a generalized autoregressive conditional heteroskedasticity model for changing variance. It points to the R package fGarch, where a specified ARMA–GARCH…
The document describes a backtest for a support vector machine that classifies the next day’s movement in Bitcoin, Ethereum, and Litecoin as buy, hold, or sell. The author says the simulation accounts for fees, spread, and market depth, and reports a…
The document raises a preprocessing issue in a high-low range forecasting model that uses a hyperbolic tangent state transition. Its outputs are the first nine state-vector elements, so predictions are constrained to the interval from −1 to 1. The forecast…
The document raises a methodological question about monitoring how groups of stocks change over time. The author has stock price data and wants to cluster stocks with similar characteristics, then follow the evolution of those clusters. They report that…
The document asks how to trade a model’s frequent, short-horizon stock predictions when the predicted moves are smaller than the bid–ask spread. The author reports a model that predicts the direction of a move over the next two minutes with 75% accuracy for…
The document describes an approach to historical and hypothetical stress testing of equity volatility surfaces when the available surface history is short and lacks the crisis period of interest. The author fits an arbitrage-free SVI parameterization to…
The note explains pricing errors in the Fama–MacBeth approach to evaluating a factor model, using the Fama–French five-factor model as context. In each period, a cross-sectional regression relates assets’ excess returns to their estimated factor exposures…
The document explains why a multi-step GARCH volatility forecast may rise or fall as the forecast horizon extends. For a GARCH(1,1) model, the conditional expectation of future variance can be written recursively using the model’s persistence, the latest…
The document outlines an implementation question about hierarchical principal component analysis for financial predictors divided into five groups. It contrasts a standard PCA on the full training matrix with a two-stage idea drawn from a…
The document describes a proposed study of whether principal components extracted from macroeconomic variables can help predict stock returns. The researcher plans to regress returns on a small number of components and examine several stock styles, including…
The document asks how to estimate the mean-reversion parameters of Ornstein–Uhlenbeck and Cox–Ingersoll–Ross style processes from a state-space model. It distinguishes the constant-volatility OU process from a process whose volatility depends on the state,…
The document asks how to apply a feature transformation from financial machine learning: cumulatively sum a series, test fractionally differentiated versions over a range of differentiation orders, and choose the least order that passes an augmented…
The document considers whether hedge fund returns can be explained by static exposures to common investment factors. It warns that a high in-sample fit, such as a strong R-squared from a lasso regression, may be misleading when many candidate factors allow…
The document argues that a machine-learning trading model has no universal target success rate. The proportion of winning trades matters only in relation to the sizes of wins and losses, the trading horizon, and the strategy’s payoff structure. A…
The document addresses a practical question about simulation output from the R package fGarch. The asker wants more than the simulated observations: they also need the conditional variance and the series of errors produced by a GARCH process. The answer…
The document raises the practical relevance of extensions to Markowitz mean-variance analysis. It notes familiar criticisms of the basic framework and asks whether adding features such as stochastic volatility, transaction costs, a more realistic consumption…