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.
4,194 documents
The document clarifies how to calculate variance for daily Fama-French factor data aggregated over a month. Because factors such as SMB are already expressed as portfolio return differences, they are returns in their own right; there is no need to calculate…
The discussion asks whether a general market model could predict how long a trading strategy’s alpha will persist. It presents several perspectives: asset-pricing models can relate expected returns to risk factors, while evolutionary economics frames…
The document asks how to obtain semiannual or annual Fama–French three-factor values from the public historical data provided at daily, weekly, and monthly frequencies. Its accepted response recommends aggregating monthly factor returns to form annual…
The document considers whether stocks must be sorted into portfolios when estimating Fama–French factor models. With only three companies available, the accepted answer recommends running separate regressions on each stock’s excess returns using the factor…
Factor models explain shared movements across assets with a smaller set of common drivers. Because stocks often move together, treating their returns as independent can overstate the diversification benefit in a portfolio. A factor representation captures…
The document discusses how to compare CAPM with the Fama–French three-factor model when studying Swedish industrial stocks. It raises a design concern: regressions on individual stocks may be less standard than tests using portfolios. It also notes that…
This discussion distinguishes named factors, defined by observable economic or financial variables, from statistical factors extracted from asset returns with principal component analysis (PCA). The first PCA component often resembles a broad market factor,…
The document explains why a weighted least squares regression may multiply each observation by the square root of a company’s market capitalization in performance attribution. The key condition is a particular error model: residual variance must rise in…
The document discusses missing quarterly Compustat inputs when calculating Tobin’s Q and cautions against replacing unavailable values with zero. Missingness can reflect the underlying data model: for example, current asset fields may not apply to banks and…
The document discusses how to compare two mutual fund portfolios when their estimated alphas are statistically insignificant, even though one appears to outperform the other. Its answer cautions that an apparent performance difference may reflect different…
The document sets out a proposed simulation of monthly stock returns using a one-factor model. Asset excess returns are described as a mispricing term plus factor exposure times factor excess returns, with an idiosyncratic noise vector. The scenario sets…
The document addresses two questions about the Shanken correction for errors-in-variables in asset-pricing tests. First, it asks how the correction's covariance terms can be added when the factor covariance is a matrix, and whether multifactor models require…
The document describes a portfolio construction approach that uses ordinal rankings instead of explicit expected return estimates. In a portfolio of sorts, the centroid of each ranked group supplies the inputs, while the optimization retains a…
The second step of the Fama–MacBeth procedure runs a separate cross-sectional regression for each time period. For each month, stock returns are the dependent variable and the factor beta estimates from the first-stage time-series regressions are the…
The document compares two approaches to estimating factor returns: cross-sectional regression using asset returns and factor exposures, and sorting assets by a characteristic to form a long-short portfolio. The answer says these approaches do not necessarily…
The document asks how the Fama–MacBeth procedure accounts for uncertainty in its final average coefficient. The procedure estimates regressions across periods and averages the resulting coefficients; the question is whether inference based on variation among…
The document outlines a common design for quantitative fundamental equity research. Investors select company measures spanning value, momentum, quality, and growth, rank stocks on those measures, and group them into deciles. They then choose a portfolio…
The document contrasts a two-step cross-sectional asset-pricing regression with a proposed Fama-MacBeth procedure using daily returns and monthly estimates. In the conventional setup described, time-series regressions estimate each asset’s factor loadings,…
The document explains why the Fama–French three-factor time-series model uses returns on size and value factor portfolios instead of inserting a company’s market capitalization directly into its return equation. The portfolio return acts as a…
The document explains how to separate an industry factor’s returns from broad market movements when estimating security exposures. It describes a Gram–Schmidt-style adjustment: regress industry returns on market returns, subtract the market component using…
The document considers whether a strong balance sheet, especially low debt or substantial cash, can give a company a durable competitive advantage. The proposed mechanism is financial flexibility: cash reserves may help a firm fund timely investments and…
The document distinguishes cross-sectional momentum (CSM), which ranks assets against peers, from time-series momentum (TSM), which measures each asset’s own absolute performance. The discussion presents CSM as useful for isolating relative, asset-specific…
The document explains why the characteristic portfolio for a chosen asset characteristic is not constrained to have weights summing to one. In the cited construction, the portfolio minimizes variance subject to achieving a unit exposure to the…
The document considers whether widespread index investing can push individual stocks away from fair value when index funds buy or sell entire baskets. It asks whether traders might exploit those distortions in the short run and who could benefit if passive…