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
This document describes an expert advisor that trades signals from the MAOscillatorHist indicator. It enters a trade when the indicator histogram changes direction, with the signal evaluated at the close of a bar. The advisor depends on a compiled indicator…
The article argues that trading researchers should judge machine learning papers by simplicity, reproducibility, and generality, alongside reported empirical results. A method that performs well on a benchmark may not transfer to financial prediction, and a…
This overview argues that no programming language is best for every algorithmic trading system. The choice depends on strategy frequency and volume, data needs, performance targets, development and maintenance costs, and the functions the system must…
This study compares ARIMA, a multilayer perceptron, an LSTM, and an ARIMA-GARCH combination for next-day NVIDIA share-price forecasts. It describes ARIMA as a model for linear time-series structure, neural networks as ways to capture nonlinear patterns, and…
This A-share stock screen combines a 14-period RSI below 65 with an external-to-internal trading volume ratio above 1.3 and positive net buying by major participants during the opening auction. The page also gives example implementations and adds filters in…
This Chinese equities screening idea combines several filters: exclude stocks marked ST, require an amplitude above one, apply a five-session closing-price high condition, and select stocks whose intraday low is down within a narrow four-to-five percent band…
The report proposes improving traditional liquidity factors for Chinese A-share stock selection by incorporating information about intraday trading structure. It frames the approach as a reusable factor research method: identify market information omitted by…
This report presents a short backtest of a strategy labeled “momentum-news-generic” against SPY, using Yahoo data. Over the stated period, the strategy had a slightly negative total return, negative annualized return, negative Sharpe and Sortino ratios, and…
This report develops a framework for rotating between the CSI 300 and ChiNext indices, whose valuations, market capitalizations, and industry exposures differ. It organizes potential drivers of their relative performance into factors covering real economic…
This quickstart walks through a bar-based exponential moving average crossover strategy in a backtesting engine. The strategy waits for its fast and slow averages to initialize, then buys when the fast average is at or above the slow one and sells when it is…
This beginner’s post describes an attempted reproduction of a quantitative research report’s “Water Drops Wear Away Stone” factor and a single-factor stock-selection backtest. The author says the implementation was prepared as a learning exercise with help…
This stock-selection idea combines three prior-day conditions: price amplitude above 1%, appearance on a trading leaderboard, and no limit-up close. The text interprets a larger range as a sign of short-term volatility and leaderboard presence as a possible…
This example script demonstrates how to connect Backtrader to OANDA data and optionally use an OANDA broker. It defines a sample strategy that logs market data and a simple moving average, tracks data and order notifications, and can submit sample buy or…
This Chinese-market stock screen combines turnover, a reversal-style candle condition, and a daily price-decline band. It selects stocks with turnover between 3% and 12%, applies a reversal filter, and keeps candidates whose daily percentage change is…
This short research note discusses how machine learning might classify stock price patterns. It uses a triple-bottom formation to show why chart labels can be ambiguous: after a decline and consolidation with three lows, an upward break may confirm a…
This document describes a Chinese equity strategy research approach that uses supervised neural networks to generate many weak alpha factors from a set of input factors. It proposes transforming those outputs into low-correlation components, then weighting…
The indicator compares the current regression slope of closing prices with stored slopes from confirmed pivot highs and lows. A pivot is identified using highs or lows across a configurable window; the slope over the bars leading into each pivot is saved in…
This research summary examines whether trading volume, represented by turnover, can distinguish stronger from weaker stock momentum signals. It separates intraday price changes from overnight returns and reports different price-volume patterns for each.…
This Expert Advisor is described as trading changes in the direction of a custom Triple Exponential Moving Average indicator. It forms a signal when a bar closes and the indicator changes direction. Running the advisor requires a separately compiled…
The author outlines an index-enhancement approach for Chinese large-cap and blue-chip equities. First, a human-selected universe emphasizes historically strong companies and stocks that held up against the broad market during declines. A machine-learning…
This Expert Advisor takes signals from the ColorSchaffRVITrendCycle oscillator when it crosses into or out of overbought or oversold territory, with the signal confirmed at bar close. The description therefore presents a threshold-break strategy based on…
This stock screen selects equities whose price range exceeds a threshold, that were not limit-up stocks on the prior session, and whose prior-day turnover measure exceeds 0.26. The article interprets elevated turnover as a possible sign of investor attention…
This equity screen combines a daily price-range filter, at least two limit-up sessions during a 500-day lookback, and a three-candle morning-star reversal pattern. The pattern is described as a long bearish candle, a small-bodied middle candle, and a bullish…
This screen selects equities using three conditions: daily price range above a stated threshold, a 2019 dividend ratio above 25%, and a nonempty name for an outstanding convertible bond. The article presents the combination as a way to consider price…