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 the partial autocorrelation function can be expressed in closed form for a Markov regime-switching process whose observations have state-dependent means and variances. Regime probabilities evolve through a transition matrix, so the…
The document describes using a logistic transformation to keep time-varying parameters within specified lower and upper bounds during maximum likelihood estimation. The proposed mapping converts an unconstrained parameter into a value between those limits,…
The document is an introductory discussion for a programmer who wants to enter quantitative finance and build relevant understanding or side projects. The answer distinguishes two possible roles: algorithmic traders, who design systems that trade…
The document considers whether the variance of an outcome should be treated as constant or modeled as changing with context. Its example is predicting NBA fantasy points: a model already estimates the expected score, and the author wants a corresponding…
The document asks how to obtain closing prices for thousands of US-listed stocks on two dates in order to calculate annual returns for a machine-learning project. The desired data covers exchange-listed equities rather than options, commodities, or indexes,…
The document asks for an introductory book on programming quantitative finance, preferably using R. The sole recommendation is a book on simulation and inference for stochastic differential equations that includes R examples. This points beginners toward a…
The document shows how to turn daily stock probability rankings into an equal-weighted portfolio that holds the five highest-ranked names and rebalances every ten days. The example uses a table of dates, stock names, returns, and model probabilities. The…
The document raises a practical calibration problem for the Heston stochastic volatility model. The author fits the model in Matlab using Monte Carlo simulations, with 5,000 paths, two time steps per day, and a 360-day horizon. Calibration with the…
The document presents a question about forecasting a differenced index series with an ARMA(2,2) mean model and a GARCH variance model. It gives the questioner’s proposed equations, a rugarch specification, and an output showing the forecast series and…
The document describes a pairs trading setup that estimates a hedge ratio and intercept as time-varying states in a Kalman filter. Its observation equation relates one asset to another, with measurement noise, and the filter updates the relationship as new…
The document compares R, Python, and Scala as tools for financial machine learning, with particular attention to unsupervised methods. One answer favors R for in-memory numeric or categorical data and exploratory algorithm development, Scala for data too…
The document explains how GARCH can be used to model changing error variance within a broader model for the relationship between an outcome and its predictors. Instead of assuming independent, identically distributed errors, a regression can allow its…
The document questions whether linear correlation is sufficient when studying relationships among financial variables and combining volatility forecasts. It describes a proposed comparison of GARCH, support vector regression, and fractionally differenced…
The document raises a terminology question about Monte Carlo estimates in reinforcement learning. It contrasts the familiar result that the standard deviation of an average of independent, finite-variance samples shrinks in proportion to the inverse square…
The document surveys ways to simplify Black–Scholes calculations, with a particularly compact approximation for short-dated, at-the-money options. Expanding the normal distribution function around zero gives an option value proportional to spot, volatility,…
The document raises the question of whether vector autoregression and related autoregressive models are useful for daily log returns in liquid financial markets. It contrasts the common use of these models in financial time-series analysis with practitioner…
The document asks how to estimate minute-by-minute premarket volume when a data source reports prices for each one-minute bar but sets those bars’ volumes to zero, while providing only an aggregate premarket total. It describes a simple allocation heuristic:…
The document gathers recommendations for readers seeking stronger statistical foundations for financial data analysis. Suggested books span risk and asset allocation, statistical learning, financial engineering, time series, Bayesian econometrics,…
The document proposes detecting temporal leakage by changing all input rows after a chosen cutoff and rerunning feature construction. If outputs at or before the cutoff change, the feature pipeline has used future data. This empirical check can reveal…
The document asks how trading firms generate research ideas before examining data, and whether they rely heavily on data mining. The question mentions arbitrage, pairs, mean reversion, tick data, momentum, and factor models, while noting practical data…
The discussion considers how an opaque trading service might produce buy and sell signals. It offers no verified description of the service’s algorithm: its own marketing claims only that the signals come from proprietary mathematical logic. The answer…
This document describes a Pine Script library for building and running machine learning models. Its primitives include tensors and matrix operations, differentiable computation graphs, neural network layers, recurrent and attention operations, and sequential…
This indicator implements a small online neural network that produces long-or-cash signals from six daily market features. The features describe relative moving-average level, RSI-based slope, distance from a regression mean, directional efficiency,…
The indicator estimates a range of possible future prices by simulating paths from recent log returns. At a user-selected chart time, it uses a rolling historical window to calculate the average and standard deviation of returns, then generates simulated…