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 collects possible sources for historical intraday market data when a terminal’s lookback limits are unsuitable. Suggestions include a tick history archive, specialist trade and quote vendors, downloadable minute interval data, exchange or…
The document describes a proposed local exchange for algorithms sharing a computer. The envisioned system would maintain virtual accounts, check available balances, route eligible orders through broker gateways, and match compatible orders internally. Such a…
The document considers whether a price crossing a moving average can avoid repeated trades during sideways, choppy conditions while still capturing sustained trends. It explains the underlying trade-off: trend-following signals tend to enter and exit late,…
This note considers the costs of buying SPY options near the market open and selling later in the day. It highlights explicit transaction charges and the bid-ask spread as separate drags on a trade. It also warns that option spreads may widen around the open…
The document asks how a continuity correction for a discretely sampled diffusion’s minimum improves the weak convergence rate of a lookback option payoff. The approximation adjusts each simulated grid value by a term proportional to the local diffusion…
The document asks how to evaluate a technical strategy across an entire stock portfolio using the Systematic Investor Toolbox in R. The example strategy enters when a lagged 50-day moving average exceeds a lagged 100-day moving average and the 25-day RSI is…
The document raises a practical issue in valuing credit default swaps across many historical dates with QuantLib’s ISDA conventions. The author wants to calculate present value, points up front, or other measures for each date, using that date’s yield curve,…
The document raises a distinction between calculating returns from weekly prices and converting daily returns into a weekly measure. For log returns, summing the daily returns across a week gives the same result as the log return computed from the week’s…
The discussion questions whether a Sharpe ratio can be estimated from many backtest runs, each covering a fixed span of trading days. The reply cautions that repeated simulations are not automatically equivalent to a historical backtest: simulated paths…
The discussion asks how to optimize trading strategies using both Sharpe ratio and maximum drawdown, especially when drawdown must stay below a chosen limit. It suggests Sortino ratio as an alternative because it focuses on downside deviation, and presents…
The document asks whether signed net volume could predict asset performance in a way that resembles the information carried by returns. It distinguishes ordinary gross volume from net volume, defined as buyer-initiated volume minus seller-initiated volume.…
The document addresses whether results from one historical period used out of sample should match results from another period used in sample after the split is reversed. Its answer is no: different periods need not produce matching estimates or performance,…
The document examines a workflow that estimates a covariance matrix from stock log returns, extracts principal components, and then projects log prices onto those components. Its central concern is that price levels are nonstationary, so the covariance…
This discussion shows how to implement upward and downward Bill Williams fractal indicators for use in R's quantstrat framework. The example identifies a local high or low by comparing a candidate bar with neighboring bars, then marks the detected pattern in…
This discussion explains how to align indicators with tick-level backtests when the indicators are usually defined on daily data. For an indicator intended to use daily observations, one approach is to aggregate the ticks into daily closes, calculate the…
The document frames a risk-management question: whether models can improve on Filtered Historical Simulation (FHS) when generating many realistic financial return paths. It describes a proposed classical workflow: select an ARMA-GARCH specification and lags,…
The document discusses how to assess a cryptocurrency portfolio rebalancing strategy based on mean reversion, especially when frequent trading costs overwhelm attractive backtest results. Its example compares hourly and daily backtests under different fee…
The document describes a forecasting workflow for Brent crude oil: transform prices into log returns, check stationarity, fit an ARMA-GARCH model, and obtain forecasts of conditional means and variances. The author reports that the forecasted conditional…
The document examines a method for estimating a 99.97% ten-day value at risk from a 99% one-day historical simulation based on 500 days of observations. It describes scaling the one-day figure by the ratio of normal-distribution quantiles for the two…
The document considers how to reduce the cost of downloading long historical price series by first screening symbols for sufficient trading history. The proposed shortcut checks a short window near the required lookback’s starting date; if a symbol traded…
The document addresses why historical daily stock data from Google Finance and Yahoo Finance may contain different dates or values. It identifies several potential sources of disagreement: different underlying historical feeds, differences in how data are…
The document proposes a regression-based method for extending a discontinued three-month commercial paper rate beyond its final observation. The first step is to identify a basket of similar interest rates with overlapping historical data before and after…
The document explains how a GARCH model represents returns and their time-varying conditional variance, addressing confusion about whether to fit the model to returns or squared returns. In the stated setup, returns equal a conditional mean plus a…
The document asks whether a portfolio optimizer is useful when its in-sample variance is a better guide to test-period variance, even though its out-of-sample variance is higher than that of a Global Minimum Variance portfolio. The example contrasts the two…