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 asks whether researchers can apply published Fama–French factors when their stock universe or screening rules differ from those used to construct the factors. One response recommends matching factor exposures to the investable universe, while…
The document compares principal component analysis and Cholesky decomposition for generating correlated normal risk factors in Monte Carlo simulation. When both methods use the same covariance matrix without reducing dimensions or altering components, they…
The document considers how to calculate an efficient frontier when assets have different return-history lengths. A complete-case approach, using only dates shared by every asset, discards earlier observations for assets with longer histories; the question…
The document poses an econometric question about comparing two long-only portfolio construction strategies tested on the same data. It contrasts testing a single strategy’s performance against the market using Jensen’s alpha with the separate problem of…
The document asks whether a portfolio’s return distribution can be formed by weighting the return distributions of its component assets. It distinguishes the straightforward weighted calculation of expected returns from portfolio variance, which also depends…
The document examines a one-period portfolio choice in which an investor allocates wealth between a risky asset and a risk-free asset under constant absolute risk aversion (CARA) utility. Its central issue is whether the amount invested in the risky asset…
The document is a beginner’s question about learning how stock markets work and finding information on shares. The reply points toward historical price information and explains that buying and selling securities requires a brokerage account. It also suggests…
The document addresses a concern that a 50/50 portfolio of two stocks shows cumulative performance above 100 percent. The response checks the adjusted-price histories individually and finds that gains of this scale can be plausible over the sample shown.…
The document explains why a portfolio’s estimated loading on the Fama–French size factor need not rank in the same order as its average company market capitalization. A factor regression measures how portfolio returns behave relative to the factor; it does…
The document asks how to combine the durations of two funds held at specified portfolio weights. It states duration as the present-value-weighted average timing of cash flows and proposes taking the weighted average of the funds’ durations. The accepted…
The document raises a portfolio optimization problem in which a variance calculation produces a negative result. It identifies the covariance matrix as a possible source: if the matrix is not positive semidefinite, some portfolio weight vector can yield a…
The document considers whether portfolio weights can be chosen by maximizing average portfolio log returns directly from observed data, without fitting a named probability distribution. One response explains that this still amounts to modeling returns: the…
The document raises a question about converting annualized arithmetic and geometric return assumptions shown in a capital market assumptions publication. It cites the familiar approximation that geometric return is approximately arithmetic return minus half…
The document asks how an asset’s mean, variance, skewness, and kurtosis contribute to its weight in a portfolio. It contrasts two possible approaches: studying how the optimal weight vector changes as moments change, or attributing weight through…
The document describes a portfolio manager’s difficulty optimizing a fixed-income portfolio using yield to maturity, duration, and credit rating. The current score favors short-maturity, highly rated US bonds when the yield curve is flat, leaving the…
The document asks whether a Bloomberg feature can plot spread duration by tenor for an issuer or sector, including measures such as DTS and option-adjusted spread duration. The eventual response says no suitable term-structure plot was available in the…
The document asks what data are needed to calculate a sector’s dividend yield from company-level cash dividends and security identifiers. The answer says to obtain each company’s spot price first, so a yield can be calculated for each constituent. It then…
The note demonstrates constructing a time series intended as a stock index from the first principal component of several stock return series. It treats stocks as variables in a principal component analysis, projects each observation onto the first component,…
The document explains why two calculations of a two-stock portfolio’s average daily return can differ. One method takes each stock’s average return and combines them using fixed portfolio weights. That calculation represents a portfolio whose weights are…
The document introduces the challenge of estimating covariance matrices for forecasting future realized volatility across assets. Covariances change over time, and estimating a matrix for many assets involves a large number of parameters, making reliable…
The discussion questions whether stronger historical returns from a concentrated index such as the Nasdaq-100 justify giving it more portfolio weight than a broad-market index. The response emphasizes sector and constituent concentration, and warns that past…
The note discusses whether finance should be treated as a subfield of economics or as a distinct discipline. It describes overlap in their history and methods: economists contributed much early work in finance, and financial research draws on econometrics,…
The document considers how a financial time-series library should calculate beta from market and asset returns. Its central design question is whether to annualize returns before calculating beta, given that practitioners use different sampling intervals,…
The document frames constrained random portfolio generation as sampling weights that satisfy linear equality and inequality conditions. Equality constraints can be maintained by taking steps within the null space of the equality matrix, for example through a…