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.
16,761 documents
The question considers forecasting a future multi-day return from a history of rolling returns. Because adjacent rolling windows share most of their daily observations, the inputs are strongly autocorrelated. The response says that autocorrelation alone does…
The document explains how to use QuantLib’s FixedRateBondHelper to bootstrap a discount curve from bond market data. It distinguishes the bond’s coupon, which defines its scheduled cash flows, from its quoted price, which calibrates the curve. Short-maturity…
The document discusses potential drawbacks of using deep learning to calibrate and price a full implied-volatility surface. It notes that neural methods can produce very fast calibration at application time, while training may involve substantial…
The discussion considers how to simulate returns for a portfolio containing many equities and bonds while retaining realistic dependence across assets. The initial proposal estimates historical means, standard deviations, and correlations, draws multivariate…
The document concerns how to design a simulation study comparing hierarchical risk parity with the analytical global minimum variance portfolio formula for stocks. The question highlights that results may depend on assumptions about the covariance structure…
The exchange discusses whether an asymmetric dynamic conditional correlation GARCH model can produce covariance forecasts beyond one day using R’s rmgarch package. The questioner believes the package supports only one-day forecasts and asks how to obtain an…
The discussion considers how to convert assets denominated in different currencies into an account currency during portfolio backtests. The central tradeoff is computational cost versus fidelity: applying an FX rate to every market tick may be expensive,…
The document outlines a basic Monte Carlo procedure for estimating single-asset Value at Risk: choose a return distribution, estimate its parameters, simulate returns, revalue the position under each scenario, and read the loss quantile for the chosen…
The note outlines a process for building a UK equity sample for Fama–French five-factor or q-factor asset-pricing research. It favors Worldscope UK over combining the active UK securities list with the dead-stock list because accounting data is more…
The discussion examines how to tell whether a market-timing strategy adds skill beyond exposure to broad market returns. A strategy that is mostly long during a rising market may appear successful while offering little independent evidence of an edge. One…
The document considers ways to hedge a long SPY position against a substantial decline: a volatility-linked ETF, an inverse leveraged ETF, and put options. It sketches possible gains and costs for each, but the proposed payoff estimates are uncertain and are…
The document discusses quality checks for daily stock records containing open, high, low, close, adjusted close, and volume. Beyond basic consistency rules—such as checking whether the close or open falls outside the daily high-low range—it recommends…
The document discusses how to represent brokerage fees, bid-ask effects, and slippage when simulating daily stock trades without quote data. It argues that a single relative transaction-cost rate cannot be assumed realistic across all situations: costs vary…
The document asks when Monte Carlo discretization is needed in financial modeling, given that geometric Brownian motion can be simulated exactly at selected dates. The responses name the LIBOR market model, Heston stochastic volatility, and local volatility…
The document discusses ways to reduce sampling noise when simulating a Cox–Ingersoll–Ross interest rate process for option valuation. It recommends quasi-random sequences, with Niederreiter sequences given as an example, and control variates, suggesting a…
The document asks how to define a stock that fades through the day in a way that can be measured. An open-to-close decline alone may miss a rise earlier in the session followed by a sharp late drop. The questioner suggests requiring an early high of day that…
The document raises two questions about reproducing a hidden Markov model study of factor investing across market regimes. First, it asks whether the reported regime labels and their return and volatility summaries use future observations to classify past…
The document discusses reducing Monte Carlo pricing variance for a European put under a stochastic volatility simulation. The questioner uses the terminal stock price as a control variate and reports a modest reduction in the estimated standard error after…
The document compares a European call price computed from a binomial model with a Monte Carlo estimate. In a binomial model, the risk-neutral probability is derived from the up and down factors and the risk-free rate. The terminal payoff is then averaged…
The document describes a simulation problem: a researcher wants to compare a trading model’s behavior on ordinary price paths and paths with jumps, while isolating the effect of increased skewness. They can match the mean of the jump process to a geometric…
The document describes simulations using the Gibbons–Ross–Shanken test to evaluate a CAPM-style model across multiple test portfolios. With returns generated under the model, the test does not reject the null; adding nonzero portfolio alphas leads to…
The document discusses how to estimate execution slippage for a strategy trading many stocks with orders that are small relative to average daily volume. One proposed starting point is a cross-sectional regression using bid-ask spread and order size relative…
The document examines an apparent performance bias in a simple rolling walk-forward backtest. It uses lagged indicators to predict next-day returns, fits a linear regression on each training window, and evaluates it on the adjacent test window. The accepted…
The document explains what it means when a test statistic falls within the non-rejection region at a stated confidence level. The result is a failure to reject the null hypothesis; it does not establish that the null is true. A non-rejection may reflect…