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 presents a discrete-time stochastic model for an exchange rate within a target zone. Its components include a drift linked to the interest-rate differential, a reference rate, a coefficient representing central-bank leaning against exchange-rate…
The document raises the practical problem of deciding when a strategy has lost its edge and should be retired. It suggests two possible warning thresholds: exceeding the strategy’s maximum drawdown or reaching a net absolute loss. These are candidate stop…
The document asks how to identify robust parameter regions in a discrete strategy grid when each parameter pair has an associated return and volatility. It proposes filtering grid points by minimum return, maximum volatility, and separate stability-radius…
The document examines a proposed position-sizing rule that scales exposure using portfolio capital, a target volatility, instrument volatility and price, and a portfolio-level multiplier. That multiplier is based on the ratio of target volatility to lagged…
A stationary process has statistical properties that remain stable under shifts in time. Weak stationarity requires a constant mean and variance, while strict stationarity requires the joint distribution across any finite set of observations to be unchanged…
The document considers returns modeled as volatility multiplied by random innovations, with the innovations following a hyperbolic distribution. It asks whether to estimate volatility first and then fit the innovation distribution to standardized residuals,…
The document asks whether a two-stage regression is appropriate for local government bond yields and their potential determinants when explanatory variables experience structural breaks at different times. The proposed approach groups variables with similar…
The document describes a data management problem in equity research: historical adjusted close prices change when later dividends or stock splits alter the adjustment factors. As a result, the value stored for an earlier date depends on which corporate…
The discussion explains how to apply White’s Reality Check or a permutation test to a pairs strategy. It represents trading decisions as a position vector, such as long, short, or flat, and randomly permutes that vector to generate simulated strategy…
The document presents a Python implementation of an explicit finite-difference scheme for the Black-Scholes equation, applied to a European call. It lays out a stock-price grid and a time grid, initializes terminal values with the call payoff, and steps…
The document discusses the rationale for testing whether Value at Risk exceedances are independent over time, as in Christoffersen-style backtesting. The question challenges whether this test is meaningful when a risk model might not adjust quickly enough…
The document considers appending a random sample to existing data while specifying its correlation with each existing series. It emphasizes that arbitrary correlations may be infeasible without assumptions about the joint distribution. Under a multivariate…
The document considers how to simulate an order book for testing strategies that submit market and limit orders. It recommends beginning with a simplifying assumption: the simulated strategy’s orders are small enough not to alter the price path. This can…
The document asks how to backtest a minimum-variance portfolio that is rebalanced at regular intervals while using all return observations available before each rebalance. It focuses on the training-period and rolling-window arguments in PortfolioAnalytics’…
The document asks whether a Hurst exponent above one has a valid interpretation. Under the conventional definition described, H lies between zero and one, with values below one-half associated with anti-persistence, one-half with random-walk-like behavior,…
The document discusses the ANTICOR portfolio algorithm, introduced in a paper on learning to outperform the best stock. It characterizes the method as a windowed form of constant rebalanced portfolio allocation and notes that its appeal rests on strong…
The document examines a Monte Carlo attempt to remove the drift from geometric Brownian motion by multiplying each simulated price path by a Radon–Nikodym derivative. Its central explanation is that this derivative changes the probability weights assigned to…
The document considers how to simulate asset paths under SABR when volatility and vol of volatility are high. It describes a Monte Carlo comparison between a call value obtained from Hagan's implied volatility approximation and an estimate from a discretized…
The document concerns building a survivorship-aware historical equity database for strategy simulations. The question is how to identify every security that has belonged to an index when ticker symbols may be reused over time. It asks whether Bloomberg can…
The document reviews ways to convert a one-day value-at-risk estimate to a longer horizon. For models where VaR is proportional to volatility, such as a normal model or a suitably specified Student-t model, volatility and VaR are often scaled by the square…
The document is a curated list of academic papers on historical option returns and strategies. It points readers to work on buy-write portfolios, option returns, volatility and skewness, dispersion trading, expiration effects, and whether options are…
The document considers how to choose a tradable universe for statistical arbitrage that models shared factors or indices and trades residuals. It contrasts liquidity cutoffs, selecting assets by their contribution to strategy performance, and using principal…
The document clarifies how drift enters Euler simulations of Brownian motion and geometric Brownian motion. For arithmetic Brownian motion, drift is an additive parameter and does not depend on the previous value of the process. For geometric Brownian…
The document compares two ways to study annual return distributions: calculating a rolling one-year return for every trading day, or sampling one return per year from a fixed annual date. The example uses long-run S&P data and reports broadly similar central…