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.
1,414 documents
This intraday strategy combines two moving averages, MACD, RSI, and ATR to select long and short entries. Long signals require price and the faster average above the slower average, positive MACD conditions, and RSI above its buy threshold; short signals…
This price-action method seeks breakouts using recent candle highs and lows. It describes opening a long when price moves above a prior high and a short when price falls below a prior low, then placing a take-profit target and stop loss. The document…
This strategy uses a short-period Relative Strength Index to trade moves back from extreme readings. It enters long when RSI crosses upward through an oversold threshold and enters short when RSI crosses downward through an overbought threshold. Depending on…
This proposed gold strategy combines Fair Value Gaps (FVG), which represent price imbalances, with Break of Structure (BoS) signals based on breaks of recent highs or lows. It describes filtering small gaps, using a lookback to define structure, and spacing…
This XAUUSD strategy combines short-term EMA direction with confirmation from a higher timeframe. It checks whether the 9-period EMA is above or below the 15-period EMA on both five-minute and fifteen-minute charts, then requires a rejection candle in the…
This Pine strategy describes a long-only grid for XAU. It divides a fixed price range into configurable levels spaced either geometrically or arithmetically. The system buys as price crosses levels downward and sells as price crosses upward to the next…
This pivot strategy combines swing highs and lows with a 50-period moving-average direction filter. It describes entering at pivot lows when price is above the average and pivot highs when price is below it, then reversing direction after stop-outs. The…
This example shows a long-only trend-following method for the current front-month gold futures contract. It compares a fast and a slow exponential moving average, entering when the fast average exceeds the slow average by a small tolerance and the contract…
This document compares alternative position-sizing methods for CME futures strategies selected from a baseline ranked by equal-weight validation Sharpe. Equal weighting treats every selected product alike, although futures contracts can have very different…
This notebook uses double machine learning to estimate whether futures carry has an effect on subsequent returns after adjusting for volatility, momentum, and cross-sectional carry rank. It distinguishes causal explanation from predictive performance: a…
This notebook checks whether the data can support a weekly, cross-sectional futures strategy before any model is fitted. It describes a design that ranks CME products, takes long positions in the highest-ranked contracts and short positions in the lowest,…
This notebook studies gradient boosting for cross-sectional prediction of CME futures returns. It varies tree capacity through leaf-count profiles and compares squared-error, absolute-error, and Huber objectives, which differ in how strongly extreme…
The document sets out how to construct forward-return labels for a cross-sectional futures strategy that ranks products by term structure. It distinguishes roll-adjusted prices, which are appropriate for returns across contract rolls, from raw settlement…
This document describes a CME futures dataset with hourly and daily bars, continuous front-month contracts, and two deferred tenors across several product groups. It explains the dataset’s coverage, fields, loading options, and related weekly CFTC…
This dataset guide describes a collection of continuous CME futures contracts spanning equity indexes, rates, energy, metals, currencies, agriculture, and livestock. It explains the hourly source data and derived daily frequency, multiple contract tenors,…
This guide introduces the public CFTC Commitment of Traders reports as a source of weekly futures positioning data. It distinguishes the Traders in Financial Futures report, which categorizes participants such as dealers, asset managers, and leveraged money,…
This document describes a daily ETF dataset intended as a candidate pool for momentum and cross-asset research. It covers nine categories, including US and international equities, fixed income, commodities, specialty funds, and currencies. Data is sourced…
The document explains how to retrieve weekly CFTC Commitment of Traders data for selected futures products and save each product’s history as a Parquet file. COT reports capture Tuesday positioning and are released on Friday; trader categories vary between…
The document analyzes static energy storage optimization as an option exercise problem. Variational analysis yields an implicit optimal exercise rule and shows how it responds to constraints such as carry costs and limits on cycling. It also compares…
The document introduces a team-developed system called the CTP Model for making daily investment decisions in gold and bitcoin. It starts from the constraint of having limited cash and time, with only historical daily prices available, and describes…
This study examines how classifiers that can abstain from prediction can shape trading decisions. A standard direction classifier always implies a market position, whereas a selective classifier can leave the portfolio uninvested when it declines to predict.…
The study compares the complexity of log-return series for Bitcoin, GBP/USD, gold, and natural gas. It applies Multifractal Detrended Fluctuation Analysis and Refined Composite Multiscale Sample Entropy, two approaches that quantify different aspects of…
The document studies statistical properties of electricity returns, highlighting both their high volatility and a strong mean-reverting tendency. It frames extreme price movements as a source of trading risk and focuses on measuring dependence across time…
The paper presents simulation methods for energy spot prices modeled as the exponential of two independent components: a mean-reverting Ornstein–Uhlenbeck process and a pure jump process. The jump component follows a compound Poisson process with positive…