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 question asks how to obtain the numerator in a market index dividend yield calculation, using Australian listed shares as its setting. The proposed formula divides the index’s indicated annual dividend rate by its current level. The response says to…
The document asks which quantities must be stable over time for a Gibbons–Ross–Shanken joint test of CAPM pricing errors. The setup uses time-series regressions of each asset’s excess return on a factor, then tests whether all estimated intercepts are zero.…
The document considers how to transform quarterly macroeconomic predictors when regressing quarterly index returns on inflation, interest rates, unemployment, and exchange rates. The accepted response recommends considering percentage changes for variables…
The document explains the timing difference in the Fama–French stock sorts: size is measured using June market capitalization, while book-to-market uses accounting equity from the prior fiscal year and market capitalization from the prior December. The…
The discussion outlines a project to build geometric-average indices for momentum and value stocks, then compare their behavior using correlations, volume and price-range measures, momentum, and moving averages. It weighs Python and R as tools for collecting…
The document asks whether high dividend yield stocks generally outperform the broad market and presents limited, mixed evidence. One cited study found that a portfolio of high-yield stocks outperformed much of the time in the British market during 1994–2007,…
The document discusses symmetric orthogonalization as a way to turn correlated equity risk factors into uncorrelated components while seeking to preserve the original factors’ interpretability. It describes constructing a transformation from the inverse…
The document asks whether portfolio allocation can avoid estimates of return means, variances, and covariances. One long response frames the issue through statistical estimation, decision theory, and investor utility, arguing that applied allocation…
The document explores how momentum results can change with the definition of the signal. It seeks alternatives to conventional lookback price appreciation, including measures based on changes between price averages, a physics-inspired construct, and returns…
The document clarifies the role of an alpha model in quantitative equity investing. In the cited investment-process description, the model forecasts excess returns for stocks. The answer treats this as a forecast of expected return relative to a benchmark or…
The document introduces Roll’s critique of empirical CAPM tests. It explains that the CAPM equation follows mathematically when a portfolio is mean-variance efficient, so testing the equation with a chosen market proxy amounts to testing that proxy’s…
The document asks why researchers often regress stock returns on Fama–French factors monthly instead of using annual observations, despite the availability of annual factor data. The answer notes that the library also provides monthly or daily data for some…
The document addresses whether cross-sectional asset pricing models such as CAPM and the Fama-French three-factor model require nominal or inflation-adjusted returns. Its central point is that either convention can be used when comparing assets over the same…
The document lays out two ways to examine a factor model such as the CAPM. In a time-series approach, each asset’s returns are regressed on a factor, and the model implies that each asset-specific intercept should be zero. In a cross-sectional approach,…
The document examines whether “alpha” has the same meaning in academic finance and practitioner portfolio management. It presents a linear factor model in which an asset’s excess return is decomposed into an intercept, factor exposures and returns, and an…
The discussion asks whether the Fama–French size and value factors remain effective beyond the sample used in their original work. It points to international evidence from Fama and French examining value and momentum across four regions. That study reports…
The document explains how to interpret coefficients from a regression of a company’s or portfolio’s excess return on the five Fama–French factors. Each beta describes exposure to its corresponding factor within the specified regression: a negative SMB…
The document derives the exposure properties of style factor portfolios formed from a cross-sectional ordinary least squares regression. Asset returns are regressed on a matrix whose columns contain the assets’ style loadings. The estimated factor returns…
The document addresses how to reproduce an international equity analysis that reports returns in US dollars and measures excess returns over US Treasury bills. Its example concerns Danish stocks and a local-currency index, both denominated in Danish kroner.…
The document asks whether the sign of a stochastic discount factor (SDF) loading can add economic meaning when testing a newly proposed asset-pricing factor. Its example is a company controversy factor, with the question of how positive and negative loadings…
The document explains how to recreate portfolio sorts associated with the Fama–French study of characteristics, covariances, and average returns. It first describes forming nine portfolios by independently sorting stocks on size and book-to-market ratio,…
The document considers ways to classify and compare an equity fund universe. Suggested approaches include estimating fund exposures through regressions against style or sector indices, using tracking error or benchmark regression to identify funds that…
The document explains why separate Carhart four-factor alpha estimates do not, by themselves, determine which of two strategies will earn higher returns. Realized performance also depends on each strategy’s exposures to market, size, value, and momentum…
The document explains a T+H lag in factor forecasting: when characteristics observed at time T are matched with returns over a later horizon H, the newest labeled training outcomes are stale by roughly that horizon. In fast-changing markets, this can make…