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.
5,922 documents
The document considers how to identify which candidate factor returns help explain mutual fund returns. It presents factor selection as a feature-selection problem and lists several approaches: rank factors using individual correlations, use recursive…
The document asks how Barra risk models construct covariance matrices from volatility and correlation estimates. It identifies exponential decay through half-life parameters and asks how Newey–West lag adjustments and a separate volatility-regime half-life…
The discussion addresses whether stock returns and estimated factor exposures should be combined for principal component analysis, and whether each dataset should be standardized first. The replies emphasize that the answer depends on the intended model,…
The discussion offers two ways to quantify underdiversification in a portfolio allocation study. One is the inverse Herfindahl-Hirschman concentration measure: take the reciprocal of the sum of squared asset weights. It can be interpreted as an effective…
The document explains how to calculate the return of a long position in one stock and a short position in another over a holding period. It recommends calculating each asset’s ordinary return, then multiplying by its portfolio weight: positive for the long…
The document asks what an estimated equilibrium return on 90-day Treasury bills means when it appears as the risk-free rate in a reference portfolio review. The accepted answer interprets the figure as a long-run estimate of the policy-rate environment,…
The document explains how the variance-covariance approach estimates portfolio volatility for value at risk. It requires historical returns for each asset to estimate the variance-covariance matrix, then combines that matrix with portfolio weights to…
The document considers whether Black–Litterman portfolio analysis can use return distributions that are not normal. It separates two roles often combined in implementations: deriving implied-return priors from a market portfolio and covariance structure, and…
The document explains how to include ridge shrinkage when minimizing portfolio variance subject to weights summing to one. It distinguishes a quadratic penalty on squared weights from a linear term and shows how the penalty fits the quadratic-programming…
The discussion examines whether replacing a zero-beta stock with a risk-free asset preserves a portfolio’s expected return while changing its volatility. It distinguishes a well-diversified portfolio, where the stock’s remaining contribution is unsystematic…
The document explains why a minimum-variance portfolio need not assign all capital to the single asset with the lowest standalone variance. Portfolio risk depends on both each asset’s volatility and its covariance with the others, so combining assets can…
The document poses a probability question: whether the chance that one stock rises can be inferred from another stock’s rise probability and the correlation between their movements. It also asks whether probabilities and correlations across several stocks…
The document considers how to add an unobserved risk factor to a correlation model when its correlation with one existing asset is known. The proposed method writes the new factor as a linear regression on that asset plus a residual. Choosing the regression…
The document discusses how to organize daily stock prices and company financial statements in a database for screening and backtesting. Its main recommendation is to model instruments, trading venues, price observations, and companies as related entities, so…
The document asks how quantitative funds build factor portfolios, such as momentum strategies, while reducing unwanted industry and style exposures. The response notes that factors commonly carry exposure to other return drivers: momentum may load on…
The document poses a valuation question about an acquired subsidiary funded partly with debt and partly with equity. It gives the acquirer’s financing assumptions, a separate target capital structure and cost of equity for comparable businesses, a tax rate,…
The document distinguishes estimating a portfolio’s variance from using a covariance matrix in mean-variance optimization. A covariance matrix estimated from historical returns can be rank deficient when the number of observations is too small relative to…
The discussion explains why Sharpe and Sortino ratios are usually not the main operating measures for passive ETF managers. Since these funds aim to track an index, their risk-adjusted returns largely reflect the index itself, leaving limited scope for…
The document addresses how to estimate the maximum duration of a fixed-coupon bond when maturity is determined from a yield curve and a target duration. The duration–maturity relationship can rise and then fall for below-par bonds, making it possible for a…
The document compares ways to explain a daily rebalanced portfolio’s realized performance by its component strategies. One approach measures each strategy’s actual contribution, but that result can depend on how the other strategies performed along the path.…
The document describes a proposed way to evaluate a multi-asset portfolio using weekly returns and a time-varying matrix of portfolio weights. The intended test samples start dates, measures performance over a one-year period, rebalances at regular…
The document asks how to select a smaller representative set from a collection of currency pairs whose return series appear similar. It reports applying principal component analysis to returns and shows variance shares and loadings for a set of…
The document proposes ways to compare how stable asset allocation weights are over time, motivated by the concentration and sharp rebalancing that can arise in classical Markowitz portfolios. It describes portfolio turnover as the sum of absolute changes…
The question describes a trade list with tickers, buy dates, and sell dates, and asks how to evaluate its performance when each stock receives equal weight. The response points to Backtrader as a Python library for building and running backtests. It offers a…