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 proposes a framework that uses macroeconomic and market variables to find historical periods resembling the current environment, then identifies which asset class, investment style, or strategy performed better in those periods. Candidate inputs…
The document raises the problem of exchanging trading signals between collaborators who want to preserve some confidentiality. A signal should not be judged by its standalone predictive strength alone: its value depends on how it overlaps with the…
This discussion considers estimating the probability that a company receives a takeover bid from signals such as pre-offer price moves, trading activity, rumors, and financial ratios. One answer recommends a logit-link generalized linear model, commonly…
The document asks how to forecast volatility across a large cross-section of equities from high-frequency returns. The accepted response says that, at the time discussed, a mature theory for forecasting cross-sectional realized volatility was lacking; much…
The document responds to a request to reproduce James Simons’s quantitative trading model. It says Renaissance used hidden Markov models in its early period, but argues that the firm’s later approach is better understood as a large, coordinated research and…
The document asks what augmented data means in Bayesian estimation of the Cox–Ingersoll–Ross (CIR) process. It describes estimating the model’s parameters from observations of a square-root diffusion and gives a discrete-time approximation. The paper’s…
The document lists statistical and structural features proposed as checks on whether generated financial correlation matrices resemble empirical ones. These include a pairwise-correlation distribution shifted toward positive values, an eigenvalue spectrum…
The document addresses whether logistic regression can be used to predict the direction of tomorrow’s return and how to validate such a model. It explains that regression is commonly used for out-of-sample generalization, but randomly assigning observations…
The document describes a modeling problem: estimate the probability distribution of the next day’s change in at-the-money Black–Scholes implied volatility using current implied volatility, days to expiration, and the underlying’s percentage move. The author…
The document asks whether there is a general way to determine how much data a model needs relative to its number of free dimensions. The author is concerned that adding variables may improve apparent accuracy while also increasing the risk of curve fitting.…
The document asks how to specify a VARX-DCC-GARCH model for several return series when the goal is to estimate dynamic and contemporaneous volatility spillovers, seasonal effects, and lagged-return effects. The example uses a VAR for the conditional mean and…
The question concerns finding valid starting values for constrained maximum likelihood estimation of a GARCH volatility model in R. The response identifies a more immediate issue in the likelihood function: it overwrites the likelihood on each loop iteration…
The document asks what happens when a financial price series is logged and then differenced, using an S&P 500 closing-price series as an example. It describes the common feature-engineering step of taking the change between consecutive log prices and relates…
The document considers how to turn a regression of next-period returns on lagged returns, an exogenous variable, and a standardized data-release surprise into a basic trading strategy. It proposes interpreting the lagged-return coefficient as a trend or…
The document presents a portfolio objective combining expected return, a covariance-based risk penalty, and a third-moment co-skewness term. It asks how to choose an optimizer when the objective’s convexity is uncertain and many weight constraints may shape…
The document evaluates whether Poisson disk sampling can serve as an alternative to crude Monte Carlo or quasi-Monte Carlo for high-dimensional integration. It identifies practical obstacles in generating samples: naïve rejection sampling has quadratic cost…
The document discusses how an experienced quant might describe their work when applying for roles that emphasize data science. The job listing highlights data curation, cleaning, lifecycle management, sourcing large datasets, and machine learning on…
The document considers how to estimate a vector autoregression that includes a constant, a deterministic linear time trend, lagged variables, and cross-factor effects. The responses suggest that the model can be expressed as a multivariate linear regression…
The document raises a modeling question about predicting credit transitions, such as movement from current status to delinquency or default. It proposes estimating state-to-state probabilities with a logit model or another method that includes covariates,…
The document discusses fitting an arbitrage-free Nelson–Siegel term-structure model to observed yields with a Kalman filter and likelihood optimization. The proposed setup uses three yield-curve factors, a maturity-dependent loading matrix, a model-implied…
Historical simulation VaR requires past market observations, but new securities may lack the price and volatility histories needed to construct stress scenarios. The note presents two broad ways to address the gap: statistical completion and economically…
The document describes a workflow for searching OHLC time series for a user-defined pattern. It extracts local highs and lows, represents each turning point as a percentage change from the previous one, and asks how to locate similar sequences in another…
The document asks for examples of using large or unconventional datasets to build trading strategies across stocks, bonds, and foreign exchange. Its only concrete example is the use of location data by investment managers. The cited news story describes…
The document asks how to choose the maximum lag in Granger causality testing on several daily time series covering roughly three years. The researcher wants to identify both whether predictive relationships may exist and the delay at which they appear. They…