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 explains how to measure out-of-sample performance for a strategy that refits its parameters weekly using the prior six months of daily data, then holds its positions for the week. It recommends calculating performance from portfolio net asset…
The document asks how to use Attilio Meucci’s entropy-pooling method to build a portfolio of five stocks when the investor has a view about conditional value at risk. Its answer points to Meucci’s paper, “Fully Flexible Views: Theory & Practice,” and to code…
The document discusses Shannon’s Demon, also called volatility pumping or the rebalancing premium: repeatedly restoring a portfolio to target weights can raise its long-run logarithmic growth under particular assumptions, even when component assets have…
This note compares Pearson and Spearman correlation for assessing whether proposed economic factors relate to portfolio assets. Pearson measures linear association, while Spearman measures monotonic association based on ranks. These measures can differ…
The document addresses how to judge the quality of return forecasts when the measured correlation between predicted and realized returns varies with forecast horizon and the size of the instrument universe. It presents a benchmark-relative framework:…
The document considers portfolio construction when asset backtests cover different periods and the objective involves correlations among daily returns and portfolio drawdowns. It compares restricting optimization to the common five-year sample with a…
The document asks how to find corner portfolios for a mean-variance optimization problem with weight constraints. These portfolios mark transitions along the minimum-variance frontier: within a segment, the same assets are held and their weights change at…
The document asks for a canonical reference establishing uniqueness of the minimum variance portfolio, defined as minimizing portfolio variance under linear constraints. It gives the key mathematical condition: when the covariance matrix is positive…
The document examines whether an asset with a positive expected return must receive a positive weight in an unconstrained mean-variance portfolio. For the stated objective, the solution depends on the inverse covariance matrix multiplied by the full…
The document considers how to size trades when signals arrive at different times, positions overlap, and committed capital cannot be reused until positions close. It proposes a bid-price approximation: subtract an opportunity cost for locking capital from…
The document poses a constrained portfolio optimization problem: minimize variance while requiring portfolio weights to sum to one, achieving a specified target return, and keeping every weight nonnegative. This last condition imposes a long-only portfolio…
The document lays out practical questions involved in estimating a multi-factor Arbitrage Pricing Theory model for Swiss equities. It considers whether factors should be unexpected innovations or can instead be observed macroeconomic series such as changes…
This discussion examines how to infer expected returns that produce chosen portfolio weights under mean-variance optimization when the covariance matrix, risk aversion, and one asset’s expected return are known. It explains that directly applying the…
The document distinguishes three tasks involving many assets: constructing a valid correlation matrix, generating returns consistent with a covariance matrix, and estimating covariance from synchronized price histories. A valid correlation matrix must be…
The document considers whether changes in working capital and capital expenditures can be recovered when a balance sheet is unavailable. It says the available information may be insufficient, but that a simplified change in balance sheet can sometimes be…
The document asks how to express Kelly sizing in absolute dollars rather than as a fraction of bankroll. It points to the standard Kelly formula and argues that knowing the bankroll lets a trader translate a fractional allocation into an amount. It also…
The document discusses ways to stress test a portfolio beyond applying arbitrary market scenarios. One proposed approach first identifies the combination of risk factors to which the portfolio is most sensitive, then constructs scenarios that affect that…
The document asks how to choose portfolio weights for stocks classified as undervalued or overvalued, given simulated return scenarios. It contrasts minimizing conditional Value at Risk, also called expected shortfall, with maximizing profit for a more…
The document examines the risk-return locus for a two-asset portfolio when the assets have perfect negative correlation. It represents the assets as returns driven by the same standard normal shock with opposite signs. For portfolio weights summing to one,…
The document explains how to calculate the covariance between a factor return and a portfolio return when the portfolio is a weighted combination of factor returns. Covariance is linear in either argument, so the covariance with the portfolio can be expanded…
The document asks whether one performance score can compare entire mean-variance efficient frontiers produced by different estimation methods. It discusses using an area-based measure rather than comparing a single portfolio’s Sharpe ratio, which captures…
The document explains how to obtain factor betas for several assets against several indices when only an exponentially weighted covariance matrix is available. It connects the familiar single-factor covariance-over-variance beta to the multiple regression…
The document explains why an FX cross rate's volatility is not generally the simple result a trader might expect from the volatilities of its two component pairs. It frames the cross as exposure to a portfolio of EUR/USD and USD/GBP positions. With zero…
The document formulates a mean–variance portfolio choice problem in which moving from current weights to new weights incurs a quadratic transaction cost. It expresses the certainty-equivalent objective using expected returns, covariance risk, and risk…