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 explains sliced inverse regression (SIR) as a supervised dimension-reduction method for financial prediction. Unlike principal component analysis, which summarizes variation in predictors without using the target, SIR uses the forecast variable…
This Expert Advisor combines a random-number gate with a moving-average price relationship to decide whether to open a position. The random draw is compared with one of eight thresholds represented by poker-hand labels; if it falls below the selected…
This stock-selection rule combines three filters: daily price amplitude above 1%, data from 2021, and a reversal or engulfing-style condition. The document describes the reversal condition through comparisons of smoothed percentage changes in recent closing…
This expert advisor generates a directional signal from a selected bar’s open and close: a close above the open indicates a buy, while a close below the open indicates a sell. It evaluates signals only when a new bar appears. A reverse mode can invert the…
This forum exchange weighs tick-level data against minute-bar data for backtesting CTA strategies. One participant notes that a year of tick observations can exceed a computer’s practical capacity, limiting the period available for testing and parameter…
This Backtrader example loads a base data feed and creates a second feed at a larger timeframe, with daily, weekly, or monthly options. The larger feed can be produced by resampling or replaying the original series, loaded separately, or created through…
The author recounts developing and repeatedly revising a set of strategies, then lays out a research-to-deployment process. It begins by defining intended capital capacity, trading costs and order size, the balance between win rate and payoff, training data…
This document describes a forex indicator that compares a fast moving average with a slower moving average. It generates a buy signal when the fast average crosses above the slow average, and a sell signal when it crosses below; signals are evaluated at bar…
This design document explains how NautilusTrader captures state-changing messages in a durable, ordered event log. Each run has its own sequence numbers, entries, and manifest; the log can be inspected, verified, or replayed to rebuild cache state. Captured…
This research summary explains a monthly stock-selection factor built from news sentiment. It combines the direction of news evaluations with their volume by summing sentiment scores for each stock over the month. The factor is refined by retaining news…
A user questions a discrepancy between a stock’s reported historical prices and the value returned by a platform’s current-data function. For a Chinese-listed equity on one trading date, the user reports historical opening, closing, and low prices that…
The distance approach forms pairs by rescaling each asset’s training-period prices to a common range, calculating the sum of squared differences between each pair’s normalized series, and selecting the closest matches. In the cited original study, the…
This document outlines a simple foreign exchange Expert Advisor built around Williams Percent Range (WPR). It sells when the oscillator crosses downward through an overbought level and buys when it crosses upward through an oversold level. The signal is…
The document describes a script that writes closing prices for all symbols available on a trading platform to a CSV file. The user can choose the output filename and select which bar’s close to export, making the script a basic way to collect a cross-section…
This report overview surveys supervised learning methods with potential uses in investment research. For regression, it distinguishes penalized models such as Lasso, ridge, and elastic net, which can address correlated inputs, from nonparametric approaches…
This Chinese-language post outlines an A-share screening idea focused on metaverse stocks. Its initial filters are prior-day turnover above 8% and a measure intended to identify afternoon large-order net inflows. The author discusses dependence on…
The document describes an automated strategy that places buy-stop and sell-stop orders around the current price. It uses a configurable initial distance and checks the spread before placing orders; if the spread is above a set limit, it skips placement. The…
Firebird v0.60 is described as an expert advisor that trades using a moving average and a configurable percentage distance between the price and the average. Before opening a position, it checks the time since the previous position was opened and skips a new…
The document explains a Kagi chart variant plotted on the chosen timeframe so that the timing of line formation is visible. Its construction depends on price moving a specified distance and breaking prior local highs or lows, rather than on moving averages…
This document describes a stock screen using a daily high-low range of at least 1%, relative volume between 1.5 and 6 times a five-day average, a share price below 12, and a price-to-book ratio below 3. The stated intent is to find shares with noticeable…
The note explains how corporate actions such as stock splits, bonus shares, or ex-dividend adjustments can change a stock’s quoted price and share count without representing an equivalent loss in economic value. A backtest that compares unadjusted prices…
This documentation explains platform support for listed, crypto, spread, and binary options, including differences in their metadata and identifiers. It describes subscribing to venue-provided Greeks either for an individual contract or for a series-level…
This forum exchange discusses modeling different futures commission schedules by contract, including cases where closing a position opened the same day incurs a higher fee. It says the backtester’s basic settings handle percentage fees and describes fixed…
The document describes a long-only stock strategy that trades when the closing price is above a rising 200-period moving average. It uses a faster and a slower moving average as a crossover trigger: a bullish cross opens a position, and a bearish cross…