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.
16,761 documents
The discussion addresses how to store daily or lower-frequency economic and financial observations that may be revised after their initial release. Its central method is to keep the observation’s reference period separate from the date each version becomes…
The document frames out-of-sample strategy evaluation as a model-selection problem. It observes that an in-sample Sharpe ratio alone is a weak guide to performance on unseen data and asks practitioners to share a more reliable metric or a combined evaluation…
The document asks whether the original constituents of the Dow Jones Internet Composite Index, held from a date before the dot-com crash, ultimately delivered returns that justified their late-1990s valuations. The proposed comparison would keep the initial…
The document considers a day-trading system that learns patterns in one stock on a one-minute timeframe. The question raises the risk that intensive research on a single name could overfit its historical behavior or stop working if market sentiment changes.…
The discussion considers how often to calibrate or rebalance equity factors built for research on the S&P. It gives a factor-specific answer: momentum may be calibrated daily, while slower-moving value measures may suit monthly calibration. This implies that…
The document distinguishes two practical components of trading costs: broker commissions and slippage. Commissions are described as charges per share, while slippage reflects execution prices that differ from the market and depends on trading volume, order…
The document asks how to apply the Euler–Maruyama method to a one-dimensional stochastic differential equation with time- and state-dependent drift and diffusion. Its example starts at one and asks for a Monte Carlo estimate of the expected positive part of…
The document presents a short-sale backtest example and asks why an equity calculation differs from a return estimate based on the ratio of the mean price to the close. The trader sells short using a quantity determined by dividing available capital by the…
The document recommends resources for learning how to apply machine learning and deep learning to finance. It distinguishes between financial research practice, neural-network implementation, and data-science programming. The central recommendation is a…
The answer argues that historical tick data should be evaluated for integrity because vendors using the same exchange feeds can still deliver materially different records. It identifies packet loss and session recovery behavior, parser errors, lossy…
The document answers a question about calculating weekly turnover for a portfolio that rebalances regularly. It gives a basic measure: take the smaller of total purchases or total sales and divide by the portfolio’s market value. This convention measures one…
The document surveys sources of historical stock market data for research and model development, emphasizing coverage, cost, format, and service quality. It describes a paid vendor option with broad consolidated, OTC, and exchange coverage and API access,…
The document examines how to choose a position when forecasts provide expected returns and volatility over multiple future horizons. It starts with the growth-oriented rule of sizing in proportion to expected return divided by variance, then considers…
The document addresses confusion about the sign and absolute value of Value at Risk outputs, using two negative figures as an example. The response redirects the question from which output is inherently better to whether the underlying risk model describes…
The document surveys ways to turn value investing into measurable stock-selection signals. It lists valuation measures such as dividend yield, earnings yield, price-to-book, return on equity, and accrual-based earnings quality, alongside related signals…
The document asks whether a training set spanning different market regimes should use one global class weight or separate weights for each period. Its example is a two-year sample with a bullish first year and a bearish second year: a global adjustment for…
The document addresses how to estimate Value at Risk for a portfolio that receives regular contributions and is rebalanced between a risky asset and a risk-free bond. Its central distinction is between external cash flows and investment performance: deposits…
Financial models serve several purposes: forecasting selected quantities, pricing instruments relative to market quotes, describing economic relationships, and helping investors make decisions under uncertainty. A volatility model such as GARCH can use…
The document explains what a CSV format requiring both date and symbol columns means when supplying data for multiple securities. A wide table may place dates down rows and securities across separate columns, which makes it seem that symbols cannot also…
The document discusses how to obtain intraday financial data when many APIs provide only daily prices. It describes a historical approach using a vendor that offered a few days of tick-level bid, ask, and trade data, which could be collected over time and…
The document considers how to model dependence when individual market series have heavy-tailed, approximately Student t marginals. It explains that a conventional multivariate t distribution imposes a t copula with a shared degrees-of-freedom parameter,…
The note clarifies the timing and roles of variables in an OLS model built from principal component projections of asset returns. The dependent variable is the accumulated log return over the next H periods, from t+1 through t+H. The explanatory variables…
The document describes a trader’s concern about a strategy tested on one second snapshot data. The trader reports that results from 2018 and 2019 broadly fit expectations, while tests on 2016 and 2017 show small negative annual returns. They note two…
The document concerns a covariance matrix used in a Gaussian construction for stratified Monte Carlo sampling, motivated by simulating a barrier option under Black–Scholes dynamics. The answer identifies a normalization error: the stated result assumes the…