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 asks how to turn the output of a Johansen cointegration test into a spread for a two-stock pairs trading strategy. The trader contrasts the method with a spread formed from an ordinary least squares hedge ratio, where one stock is reduced by a…
The document describes an attempt to assess a GARCH model fitted to S&P 500 returns. It shows a rolling backtest setup and reports difficulty using mean absolute percentage error (MAPE) to compare predicted and realized values. The reply explains that MAPE…
The document considers how to balance uncertain expected-return forecasts against covariance risk and trading costs in a constrained portfolio optimization problem. Its proposed method is to multiply the return signal by a shrinkage coefficient between zero…
The document clarifies how conditional variance depends on the variable being modeled, in response to a question about comparing implied volatility with GARCH forecasts. Under information available at the prior time, a price and its simple change have the…
The document surveys commercial suites that combine market data, backtesting, event processing, and automated trading. It asks about leading vendors, practical tradeoffs, integration costs, customization, and typical users, but provides only a partial…
The document outlines how to test the CAPM by comparing asset excess returns with market excess returns through estimated betas. It explains that the Kenneth French data library supplies market returns and a risk-free rate, but generally does not provide…
The document demonstrates how to use Halton low-discrepancy sequences in Monte Carlo option pricing. Uniform sequence values are transformed through the inverse standard normal distribution, then scaled by the drift and volatility of a risk-neutral log-price…
The discussion compares drawdown objectives with minimum variance and expected shortfall in portfolio allocation. It presents several reasons drawdown may be less common: variance is a familiar target, drawdown optimization can be computationally demanding,…
The document considers whether statistically simulated price histories can reduce the risk of selecting strategies that merely fit past winners. One reply warns that simulation adds a second assumption: the market model must be credible before a strategy’s…
The document examines how to convert a daily Sharpe ratio into an annual figure. It illustrates the issue with a daily mean return and standard deviation, then compares compounding the average return over a year with scaling volatility by the square root of…
The question describes applying principal component analysis to many correlated stock features while trying to predict the FTSE 100. It raises confusion about a result suggesting that no components explain 99% of variability, whether additional features such…
The document raises a numerical issue in Monte Carlo valuation of European calls and puts on the same stock. Under standard simulation, each option value is estimated by averaging its payoff over simulated underlying paths. With a finite sample, the…
The document distinguishes almost-sure convergence of a Monte Carlo sample mean from the rate at which its estimation error typically decreases. Under the stated independent, identically distributed sampling setup with finite variance, the central limit…
The document considers how to model sovereign yield spreads for Germany, France, and Spain using unemployment and public debt. It raises choices about stationarity, whether to use absolute economic indicators or differences between countries, and how to…
The document discusses how to combine rolling-window estimation with the two-stage Fama–MacBeth method. The questioner estimates portfolio factor exposures over rolling five-year samples, producing many sets of estimates, and asks which exposure estimates…
The document describes a data-access problem involving Nasdaq Data Link’s CHRIS continuous futures datasets. A user could retrieve other datasets with a valid API key but received access-denied errors when requesting historical daily futures data intended…
The document asks how to choose a stop-loss boundary automatically from a trade log containing entry and exit times, profit or loss, maximum adverse excursion, and maximum favorable excursion. The suggested approach frames threshold selection as an…
The document describes a replication of an equity statistical-arbitrage strategy in which a stock is hedged with a basket of ETFs. Regression residuals provide the trading signal, and regularized regression estimates the ETF hedge coefficients. The…
The document considers whether to subtract the sample mean from log returns before applying a block bootstrap. Its main answer says centering is useful when the bootstrap is meant to represent a null hypothesis of zero expected excess return. Resampling…
The document explains how to organize constituent returns when researching strategies on stocks that enter and leave an index. For a simulated portfolio that follows membership, include a stock’s returns only while the strategy has exposure: add it upon…
The discussion frames the task as explaining changes in a strategy’s daily profit and loss using candidate variables that were not used to generate its trading signals. It recommends first considering how the predictors relate to one another: dimensionality…
The document frames a modeling question about predicting the difference between a bond auction’s yield and its when-issued yield. The data are described as sparse: auctions occur monthly, and the author reports 850 results across maturities over a…
The document addresses how to evaluate GARCH forecasts, which predict conditional volatility rather than the return level. It outlines an out-of-sample comparison associated with Hansen and Lunde: estimate an ARCH-family model using data through a chosen…
The document outlines several ways to assess whether a candidate signal adds useful return information. Suggested checks include risk-adjusted performance and tail risk, regression against established factor models to examine the intercept, trading costs and…