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 collects suggestions for learning quantitative finance online, with an emphasis on free or lower-cost study. The original question names material on financial engineering, option pricing, Monte Carlo methods, and MIT OpenCourseWare, while noting…
The document considers how to classify Bitcoin’s next-day percentage change into ordered rise, near-flat, and fall categories. The questioner’s initial scheme uses fixed percentage thresholds, but asks whether a given percentage move should be classified…
The document raises a time-series forecasting question about why differenced ARIMA models may be useful over short horizons while error correction models are described as more suitable for long-term forecasts. It cites the distinction between modeling…
The document discusses model choices for classifying financial time series into positive, negative, or neutral outcomes according to whether a forward return crosses a chosen threshold. The responses point to support vector machines and random forests as…
The document considers what to conclude when residuals look roughly bell-shaped in a histogram and quantile plot but fail the Jarque–Bera test. Its central advice is to treat a powered statistical rejection as evidence against normality, rather than using…
The document describes a lending operation in which clients receive advances against goods held in a warehouse and repay from later sales proceeds. If sales fall short, arrears can accumulate while clients continue borrowing, potentially leaving advances…
The document examines whether a rolling-window walk-forward backtest produces genuinely out-of-sample results. In this approach, a strategy is fitted on past data, evaluated on a later period, and periodically refitted as new observations arrive. The…
The document raises a model-design question for trading signals: should a strategy predict return direction, such as a positive or negative label, or forecast the magnitude of the return? Directional classification can be modeled with a binary response,…
The discussion asks whether F# has an established standard library for quantitative finance and what kinds of financial work such a library supports. The answers report no broadly accepted, comprehensive F#-specific package, while pointing to a varied…
The document concerns a fitted ARMA model whose residual autocorrelation plot appears to retain a large positive spike at the first lag. The response explains that autocorrelation at lag zero is always one: a series is perfectly correlated with itself, since…
The document describes a trading-modeling question in which neural networks and support vector machines show strong internal validation accuracy on three-class labels but perform near chance on a separate dataset. The models use many technical and…
The document asks whether Longstaff-Schwartz can determine when to sell a stock within a fixed time horizon, rather than being used only to value derivatives. The answer places the method within the broader class of finite-horizon optimal stopping problems,…
The document asks whether stock forecasting models can combine recent price movements with longer-term trends. It introduces ARIMA as a common time-series model and ARCH-type models as approaches to changing variance, then questions whether methods focused…
The document raises the problem of defining target variables for machine-learning models that predict stock behavior. Candidate labels include the next day’s net return and a return based on the maximum high over a future window. It also references a…
The answer offers practical checks for choosing equity data used in technical-analysis and machine-learning research. It advises excluding thinly traded stocks when recorded prices may not reflect executable prices, recognizing duplicate economic exposure…
The document discusses how statistics and machine learning overlap in quantitative research interviews. Its answers characterize statistics as the study of sampling, probability, and data distributions, with hypothesis testing and inference among the…
The document collects suggestions for learning about machine learning and data mining with an interest in trading. Recommendations include general machine learning courses, practitioner blogs, a software resource catalog, Q&A communities, and a time-series…
The document examines whether a random-walk state transition is suitable when a Kalman filter estimates dynamic weights for a cointegrated portfolio. It proposes a diagonal transition matrix in which the component associated with the stationary portfolio…
The document compares two ways to identify swing highs and lows in a price series. One Python-style method tracks the latest high or low and changes direction after price moves by a specified absolute threshold. Raising the threshold produces fewer reversal…
The author is trying to identify ARFIMA models and estimate the fractional differencing parameter through the Hurst exponent. They report inconsistent estimates across several software libraries when testing combinations of autoregressive, moving-average,…
The document asks how to normalize stock data and define rewards for a reinforcement learning trading system. The author describes transforming price changes into percentage returns, applying min-max scaling, and using the Sortino ratio as the reward. The…
The document examines whether statistical tests on weekly strategy returns remain reliable after parameters have been repeatedly adjusted to maximize annual performance. Its central point is that this selection process creates overfitting: the same sample is…
The document starts from the Markovian log-utility portfolio problem, where expected return and volatility depend on the current state. It summarizes the associated Hamilton–Jacobi–Bellman equation and gives the familiar feedback allocation proportional to…
This discussion presents a characteristic function for the Wishart stochastic volatility model associated with Heston-style pricing, with the aim of using Fourier methods. It outlines how a matrix Riccati equation can be linearized and solved from blocks of…