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.
1,116 documents
This research summary tests stock selection factors built from return and trading-volume data, using moments from the first through fourth orders. It places these measures in a broader taxonomy: factors can use one data dimension, combine dimensions at one…
This high-frequency trading note proposes tracking aggressive order flow over a rolling window of ticks. It explains the Bayesian idea as updating beliefs when new evidence arrives, then applies that intuition to counts of buy- and sell-side activity. In its…
This tutorial presents a workflow for evaluating a high-frequency grid market-making approach on Binance Futures. It covers selecting trading pairs, obtaining historical depth and trade data, converting that data into the backtester’s format, modeling…
The discussion addresses how to feed minute-frequency market data into an AI stock-ranking workflow. Its answer is to replace the stock ranker template’s basic feature extraction step with direct data-source extraction, because minute data does not come with…
The document reviews weaknesses in high-frequency and market-making approaches that place orders too close to the market. During sharp one-minute moves, nearby orders can accumulate inventory quickly, increasing stop-out risk and slippage. It suggests…
This research summary describes a short-horizon A-share selection factor built from intraday data. It calculates realized volatility, realized skewness, and realized kurtosis, then uses a regression model and the residual standard deviation from each period…
This research summary proposes using a convolutional neural network (CNN) to distinguish structured return sequences from shuffled sequences as a way to investigate weak-form market efficiency. In a simulation, pattern segments are inserted into white noise…
The document describes a tool for evaluating many equity factors in batches and saving overall and yearly long-side results. Users configure the factor source and names, stock universe, benchmark, optional smoothing, and output settings. For high-frequency…
This guide describes a VeighNa workflow for recording large volumes of futures tick data in TDengine. It outlines running TDengine in a Docker container, configuring database storage and ports, and connecting VeighNa’s CTP gateway and data-recorder app. The…
This article describes a set of purported Chinese A-share rules intended to curb speed and access advantages in quantitative trading. It says the measures would add network delay, end broker-dedicated trading positions, cap order submissions, require…
The discussion explains where users can find tick data downloaded through VeighNa’s data management module. One user reports locating the records in the database but not seeing them in the module’s graphical interface. A respondent explains that the…
This research note studies whether the timing of stock trading value within the day contains information beyond conventional turnover factors. It examines minute-level turnover distributions, a turnover-share factor over 240 minutes, higher moments, and…
The document describes a framework for computing quantitative stock factors as recursive transformations of numerical indicators. It says factor logic can be defined with programming languages or XML: procedural languages suit more complex structures, while…
This article describes a way to assess whether a feature built from recent stock price and volume data helps predict returns over the next five minutes. Using minute data for the 50 constituents of the SSE 50, it aligns feature values with future-return…
This research note examines how stock trading volume changes across the trading day and how the timing of that volume relates to subsequent returns. It describes a W-shaped intraday pattern in China, with activity rising around the morning and afternoon…
This article outlines a high-frequency strategy research framework for China’s order-driven markets, using Level 1 order book data and support vector machines (SVMs). It proposes extracting features such as best bid and ask prices and sizes, derived depth…
This short forum post recommends adding the Amihud factor to a high-frequency factor table associated with a multi-factor alpha research series. The author says the factor was omitted from the table and characterizes it as a strong long-side factor. The…
This study reproduces high-frequency stock-selection factors from brokerage research using one-minute OHLCVA data and a sample of roughly 350 constituents of the CSI 300 over a historical period ending in 2022. It focuses on five daily factors formed from…
This document describes a utility for creating synthetic order latency observations from market feed data. It keeps events that contain both exchange and local timestamps, resamples them at a configurable interval, and uses each interval’s last timestamps to…
This brief description outlines a manual scalping system that uses five indicators on a five-minute chart. It gives a color-based entry rule: consider a long position when all indicators display blue and a short position when all display red. Signals are…
The document outlines an equity factor research workflow: define a hypothesis from investment or behavioral theory, gather and clean data, calculate factor values, evaluate each factor, combine factors, and backtest and refine the result. It highlights…
The author describes a stock reversal approach based on a prior limit-up move followed by a pullback, with a bullish candle engulfing a bearish one as a defining pattern. A proposed variation focuses on a leading stock's first bearish session and subsequent…
SpreadMIN is a chart indicator that records and displays the lowest spread observed during each candle while it is running. The accompanying idea is to delay a trade until the current spread is no greater than the minimum recorded in the previous candle,…
The document compares an asynchronous script that requests market data for many cryptocurrency pairs in a local VS Code run with its behavior in a live environment. The local log shows updates arriving about a second apart, while the live log shows selected…