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 addresses how to calculate returns for the six Fama–French portfolios formed by combining size groups with book-to-market groups. It says the published portfolio returns are available from the researchers’ websites and points readers seeking to…
The document presents conflicting views on whether Markowitz modern portfolio theory remains useful in asset management. Supporters point to its continued use in wealth management and traditional funds, its familiarity to regulators, and its role in long…
The document raises a reproducibility question about covariance shrinkage toward a constant-correlation prior in R. It names two packages that offer shrinkage routines and asks why their outputs differ when applied to the same generated data. The response…
The document outlines a Monte Carlo approach to estimating portfolio value-at-risk for a portfolio with multiple assets. It starts with the portfolio’s current value and an estimated covariance matrix, then uses a matrix square root such as a Cholesky factor…
The document explains the meaning of a fund holding listed as a total return swap on the S&P 500. A total return swap exchanges cash flows tied to an underlying asset or index’s total return, incorporating both price changes and dividends. In the example,…
The answer explains Merton’s passage from discrete portfolio changes to continuous-time notation by rewriting the sums as integrals. It identifies the consumption term as an ordinary time integral, the left-endpoint sum as an Itô integral, and the sum of…
This question distinguishes time-series betas, estimated for each stock from historical returns, from cross-sectional factor exposures used to describe a portfolio’s sensitivity to common risks. Sector exposures are straightforward: use membership…
The document asks how to compare sample, single-index, industry-factor, and principal-component factor models using their implied correlation matrices and minimum-variance portfolio weights. It describes plots from an analysis and proposes that a model whose…
The document asks whether summing daily value-weighted returns, with dividends reinvested monthly, should match the corresponding monthly return. The reply explains that returns over successive periods compound multiplicatively rather than add directly. Its…
The document considers portfolio weights chosen to bring portfolio variance close to a target while meeting a minimum expected-return requirement and a fully invested budget constraint. It discusses solving the resulting optimization at each rebalance date.…
The document asks whether ex-ante active risk from asset allocation can be combined with ex-post active risk from security selection to estimate a multi-asset portfolio’s total tracking error. It raises the key issue that simply adding risk measures assumes…
The document discusses the Security Market Line (SML) under the Capital Asset Pricing Model when the risk-free rate is negative. In the CAPM equation, the risk-free rate sets the SML’s vertical intercept; lowering it shifts the intercept downward. With the…
The document considers whether removing highly correlated indexes from an equal-weight portfolio can improve diversification, and whether negatively correlated assets should also be excluded. The suggested alternative is to choose portfolio weights by…
The document raises a portfolio-hedging question: how to account for correlations between international bond returns and foreign-exchange carry returns when choosing currency hedge positions. It defines the relevant series as asset log returns and carry…
The document concerns choosing which retail credit portfolios to retain or sell when only aggregate portfolio information is available and borrower-level records are not. Its response points toward loss given default (LGD) modeling as a relevant area of…
The example asks why a portfolio beta estimated by regressing portfolio returns on market returns might not equal the weighted average of the individual asset betas. It constructs portfolio returns from two bond price series using weights based on estimated…
The discussion outlines a basic process for benchmarking a mortgage-backed securities portfolio. First, choose a relevant index, with U.S. mortgage benchmarks given as examples. Then compare portfolio data with benchmark data, taking care that the holdings…
The discussion considers whether intraday volatility and flash crashes matter to investors who hold portfolios for the long term. It distinguishes ordinary short-lived price moves from effects that can persist through trading costs, market impact, and…
The document shows how to impose limits on aggregate portfolio exposure to categories of ETFs, such as bond funds or funds in one asset class. In a quadratic-programming formulation, each category is represented by an exposure vector marking which assets…
The document explains how the risk-aversion coefficient in mean-variance utility relates to the maximum-Sharpe portfolio. Its central point is that the coefficient does not determine the maximum Sharpe ratio. With a risk-free asset available, the…
The document asks whether a mean-variance framework can make asset correlations depend on investor demand instead of treating the correlation matrix as fixed input. This matters because portfolio choices and market outcomes may influence one another: demand…
The document explains why the market portfolio in modern portfolio theory is a useful theoretical construct but not a portfolio investors can fully buy. A literal version would include assets beyond publicly traded securities, including private or otherwise…
The discussion considers how to reduce correlation among factor-mimicking portfolios built from Fama–French style sorts. Because the traditional sorting procedures do not explicitly control correlations, one proposed approach is to construct replacement…
The discussion explains why highly correlated asset returns create problems when estimating models or working with covariance matrices. Perfect collinearity makes the relevant matrix rank deficient, while near-collinearity can leave very small eigenvalues…