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.
2,013 documents
The document explains why two assets can offer portfolio benefits even when they have identical means and standard deviations. Their individual return statistics describe only the first two moments of each asset; they do not capture how the assets move…
The document describes a data-preparation problem encountered when collecting weekly closing prices for several international equity indices and funds. The returned series have different row counts despite covering the same requested date range. Different…
The document considers a security that pays the product of two asset values at maturity, where one asset is traded and the other is not. It assumes the assets follow geometric Brownian motion and are uncorrelated. The proposed approach is to specify a…
The document surveys ways to connect a Python trading strategy to brokers for sending, modifying, and closing orders. The responses mention MetaApi’s cloud SDK as a route to brokers that support MetaTrader, Interactive Brokers’ API for several asset classes,…
The document asks how to turn simulated returns from an ARMA-GARCH-EVT-copula model into allocations for three portfolio objectives: maximum Sharpe ratio, global minimum variance, and minimum CVaR. Its central practical questions are whether the simulated…
The document addresses how to obtain interest-rate yields for a country when a data platform returns currency performance instead. Its central point is that there is no single domestic risk-free rate suitable for every analysis: the analyst must choose a…
The document asks how to compare companies’ revenues and expenses in a Data Envelopment Analysis when their financial statements use different currencies. It contrasts converting amounts at direct exchange rates with using purchasing power parity (PPP)…
The document discusses whether a general time-series store should use UTC timestamps or local market time. Its main recommendation is to store timestamps in UTC and retain each series’ time-zone context so that local times can be reconstructed when needed. A…
The document distinguishes money management as the broad activity of managing an investor's assets from asset allocation and portfolio optimization as parts of that activity. It describes decisions at three levels: strategic allocation across broad asset…
The document asks how to set market capitalization weights for a Black–Litterman portfolio holding international and emerging-market equities, gold, long-term Treasury bonds, and real estate. Its central advice is not to use the market values of the ETFs…
The document asks how to extend a Monte Carlo portfolio Value at Risk calculation to holdings from different asset classes, using a stock, a Treasury bond, and an option as examples. It refers to a simulation approach involving correlated random returns and…
The document explains how to choose a representative benchmark when a trading strategy is opaque but has extensive historical data. A clear strategy, such as one holding only long positions in technology stocks, can be compared with a closely related market…
The document outlines a proposed procedure for generating observations from a Gaussian copula whose correlation matrix changes over time. It describes the Dynamic Conditional Correlation recursion for updating an intermediate matrix from its long-run value,…
The document asks how to construct a corporate revenue-to-world-GDP variable for a model of firm size. It compares current-dollar GDP, which converts contemporaneous local-currency values into USD, with constant-price GDP, which adjusts for inflation. The…
The document responds to a question about finding a website where a British investor with modest capital could compare past performance of British and overseas hedge or mutual funds, while filtering by minimum investment. Its answer focuses on why hedge fund…
The document addresses how to map financial instruments across identifiers used by data vendors and standards, including provider codes and ISINs. The answer points to commercial reference-data sources as possible ways to obtain mappings, while noting that…
The document considers how to optimize a portfolio over a short horizon when hedge fund or other illiquid asset indices are available only at low frequency. It questions whether monthly observations provide enough data for a two-year estimate and rejects…
The discussion outlines ways to separate currency effects from local-market allocation and security selection when portfolio and benchmark weights and returns are available. One approach measures currency value added with a simple currency attribution, then…
The document is a question about selecting discount curves for pricing government bonds across developed and emerging markets. It contrasts using overnight index swap rates for each currency with constructing curves from government zero-coupon rates, and…
The document describes a barrier reverse convertible linked to several underlyings, including a basket, with a coupon, a European barrier, and repayment tied to the worst-performing asset if a barrier condition is met. The proposed decomposition is a…
The document asks whether audited financial statements or other credible public evidence can verify Renaissance Technologies’ reported success. Its answer explains that a private hedge fund generally need not publish audited statements for the public, though…
The document addresses how to group countries as developed or emerging for a cross-country study of anti-corruption law adoption. It notes that there is no single official classification that all researchers must use. Instead, the choice of classification…
The document asks how investors could trade exposure to the number of beds in purpose-built student accommodation, treating bed count as the underlying rather than seeking exposure to rental income or total returns. The response says a bespoke instrument…
The document asks how portfolio construction can limit reliance on expected-return estimates, which are subjective and can make mean-variance optimization highly sensitive to inputs. Minimum-variance optimization removes expected returns from its objective,…