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 document asks whether a pairs trading strategy should retest cointegration each day using a rolling window, rather than test once during an initial formation period and then trade the selected pair for a fixed holding period. The proposed approach uses…
The document addresses uncertainty in a Monte Carlo estimate of conditional value at risk (CVaR), which depends both on estimating a tail quantile and averaging outcomes beyond that threshold. It recommends bootstrap confidence intervals: repeatedly resample…
The document raises the problem of selecting a risk-free rate when valuing historical option data. It notes that rates such as Treasury or constant-maturity yields, OIS, SOFR, LIBOR, and policy rates have all been suggested, and asks which benchmark fits…
The discussion examines how missing observations can affect comparisons between observed and model-generated time series, including running RMSE. It warns that interpolation may preserve the series’ central level while understating uncertainty, and that…
The document asks how to interpret the integral of a stochastic process's intensity over the interval between two consecutive arrivals. It contrasts the usual interpretation of integrated intensity as an expected event count over a fixed time window with a…
The document asks how to build a multi-year time series for a three-month futures contract while handling contract rolls. The author is concerned that prices are volatile on roll days and asks about interpolation or using calendar spreads to create…
The document considers the reciprocal of a stock price modeled by geometric Brownian motion (GBM). It gives the standard GBM solution and uses the moment-generating identity for a Gaussian variable to derive the reciprocal’s mean. For initial price S₀, drift…
The post questions how White’s Reality Check handles strategies that generate different numbers of trades. Its example compares one strategy with two observed returns and another with ten, then describes bootstrapping returns, taking the maximum strategy…
The discussion describes what makes options strategy backtesting more demanding than testing a simple underlying asset. A credible simulation needs historical option chains, including contract terms and market prices, alongside underlying prices used to…
The document asks how to estimate a stochastic volatility model from historical daily returns without using an implied volatility surface or computationally intensive MCMC. It specifies a latent log variance that follows an autoregressive process with…
This discussion considers whether a candle-based strategy backtest can rely on mid-prices, especially when modeling stop losses, or should use bid and ask prices to reflect the side at which a trade executes. The author reports that replacing mid-price highs…
The document compares AAPL historical data from two vendors and asks why their recorded activity differs. The examples show discrepancies in timestamps, reported trades or quotes, and prices. Replies propose several possible causes: one vendor may provide a…
The document discusses ways to reduce the runtime of repeated mean-variance portfolio optimizations, such as daily rebalancing for a universe of fewer than one hundred assets. It suggests first considering whether the work can be parallelized, then using the…
The document discusses whether Yahoo Finance data accessed through a Python package is reliable for historical daily open and close prices. The questioner distinguishes actual closing prices from adjusted closes and asks about alternatives. Responses suggest…
The document discusses a machine-learning return forecast evaluated with walk-forward validation while varying training-window length, prediction horizon, step size, and embargo. The reported out-of-sample results deteriorate as the step or embargo…
The document discusses ways to assess whether a strategy’s live profits reflect a repeatable signal or chance, especially when market volatility is large relative to the observed gains. One proposed test, attributed to Alfred Cowles, preserves the trader’s…
The document asks how to construct weekly returns from daily price data using Wednesday closing prices. When Wednesday has no valid trading price, the question proposes using the next available valid closing price. It refers to log returns between selected…
The document considers whether measuring geometric Brownian motion at weekly rather than daily intervals lowers annualized volatility. Under the stated assumption of independent, normally distributed returns, variance grows linearly with time. Scaling the…
The document asks how to include lags when applying least squares Monte Carlo to a timing real option. The project can be started at monthly decision dates over a five-year window, while revenue follows a non-Markov ARMA process sampled hourly over a…
The document addresses why historical values from Yahoo Finance’s API may differ from those shown in its interactive chart. The answer focuses on the distinction between Close and Adjusted Close: the chart may display adjusted values that account for events…
The document addresses the tedious process of downloading adjusted closing prices for several stocks individually and merging the resulting files. It presents two programmatic alternatives: using a Python data-reader library to request multiple symbols over…
The document asks whether transaction records and candlestick summaries can reveal the highest bid, lowest ask, or quote sizes in the foreign exchange market. The available data includes trade times, amounts, prices, and bar-level measures such as highs,…
The document asks how to convert dollar P&L into percentage returns for a two-leg straddle strategy sized by vega notional. The strategy aims to stay vega neutral by balancing its legs, while positions are adjusted and delta hedged over time. The central…
The document discusses how to choose the observation window and retesting frequency for cointegration methods such as the ADF and Johansen tests when designing a pairs strategy. It rejects a universal lookback standard and relates the choice to data…