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 document compares two ways to forecast a stock’s performance relative to a benchmark: model each asset’s returns separately and subtract the forecasts, or model the relative return directly. It frames the problem as a weekly time-series forecasting task…
The document describes an apparent discontinuity in weekly historical prices retrieved through quantmod for an Indian equity. The response traces the discrepancy to the underlying Yahoo Finance history rather than to quantmod itself: the same inconsistent…
The document asks how to backtest a stock-direction model that is retrained on historical price windows and used to generate trades. The author reports prediction accuracy and wants to test a strategy over multiple years, including risk controls such as stop…
The document explains how to obtain one-step-ahead conditional mean and volatility forecasts across a time series after fitting an sGARCH model with rugarch. A standard call to the forecast function returns a forecast at the end of the fitted sample, which…
The document describes a continuous futures series built by adding a cumulative roll adjustment to settlement prices and applying the contract multiplier. With this additive adjustment, the resulting notional can become negative even when the underlying…
The document addresses poor Monte Carlo pricing accuracy for deep out-of-the-money options under a stochastic volatility model. The question focuses on Euler simulation of the Heston process and asks whether importance sampling can reduce variance and speed…
The document considers how to assess long-volatility strategies with frequent small losses and occasional large gains, where mean return, Sharpe ratio, and hit rate may not capture the payoff pattern. The response recommends beginning with a long history of…
The document explains why cumulative returns depend on how profits and losses affect the capital base. If gains and losses remain invested and change the amount exposed to later returns, calculate the total by multiplying each period’s gross return and…
The discussion looks for free historical end-of-day prices for US stocks, especially S&P 500 constituents, with a database that avoids survivorship bias. It mentions several paid databases and notes that Yahoo Finance access remained possible through an R…
This explanation clarifies why Cover’s universal portfolio does not generally remain equally weighted after its initial allocation. It starts with equal weights, then estimates the next allocation as a performance-weighted average of constant-rebalanced…
The document turns a set of bets with different success probabilities and payouts into a portfolio-optimization problem. It proposes minimizing portfolio variance using the covariance matrix of the investments, with each bet’s expected value and variance…
The question asks how to calculate returns and a Sharpe ratio for a systematic options strategy that may open positions with no net premium paid. The response emphasizes that zero initial cost does not mean the strategy requires no capital: the trader must…
The document raises questions about why daily and monthly portfolio return plots derived from a data provider’s return index might differ. It describes calculating daily returns as consecutive index changes and monthly returns from month-end index values. It…
The document outlines a filtered historical simulation approach to market risk for vanilla swap portfolios. Historical observations should be represented as changes in selected curve risk factors, such as discounting and forwarding zero rates at chosen…
The document asks whether monthly factor returns should be aggregated into overlapping quarterly, semiannual, or annual returns before regressing them on lagged market- and funding-liquidity variables. The proposed motivation is that monthly observations may…
The document describes a long–short factor backtest that ranks assets into top and bottom quintiles. The signal has positive performance at monthly, quarterly, and annual rebalancing frequencies, with returns improving as rebalancing becomes less frequent.…
The document considers how to reduce the worst losses of a bullish call spread by adding a bearish option position in a later expiry. It discusses balancing hedge size, the hedge’s cost, and the different rates of time decay across expiries. The proposed…
The discussion addresses why a request for index history beginning in 1960 returned data only from 2010 onward. Its practical recommendation is to locate a source that covers the missing period, preserve that historical segment locally, and periodically…
The document presents a European call pricing comparison under the Heston stochastic volatility model. It gives a characteristic-function integral for the closed-form price and a Milstein-style discretization for simulating variance and log spot, then…
The discussion addresses two distinct validation needs for mean–variance portfolio allocation. One reply challenges the assumptions behind mean–variance finance, arguing that return distributions may be heavy-tailed and that sample means and variances may…
The document considers how to compare approaches intended to address parameter uncertainty in portfolio optimization, including Bayesian methods with diffuse, asset-pricing-based, or economic-objective priors. Its response recommends evaluating forecasts…
The document asks how to apply Fama–MacBeth regressions to monthly asset returns when each asset’s predictor is its own lagged return. For each month, the proposed procedure runs a cross-sectional regression of that month’s returns on the corresponding…
The document discusses the provenance of Quandl Open Data, also called Wiki Data, and asks whether the data are purchased from exchanges or derived by comparing and transforming other free sources. It presents the dataset as a possible testbed for…
The document distinguishes GARCH forecasts from realized volatility (RV) measures. GARCH models use discrete observations, commonly squared returns, to estimate conditional variance and forecast volatility. RV instead aggregates high-frequency observations…