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.
773 documents
The article presents a notebook-based workflow for quantitative research: obtain exchange candlestick history through an API, store and inspect it with pandas, plot price and trade-flow measures, and build a Python backtest for multiple spot or perpetual…
This tutorial compares three ways to train an XGBoost model for stock selection: ranking securities by a score, classifying outcomes into categories, and predicting a numeric target through regression. It frames these choices within a broader modeling…
The document explains the Crank–Nicolson implicit finite-difference scheme for solving the one-dimensional heat equation. It contrasts this approach with an explicit method that requires small time steps, describing Crank–Nicolson as averaging spatial…
The document outlines a Chinese stock screen using three filters: a daily price-change measure greater than one percent, a listing age exceeding one year, and a closing price below 12 yuan. It presents the filters as a way to find active, established,…
This stock screen selects shares whose daily high-low amplitude exceeds 1%, whose price is above its five-day moving average, and whose 2021 revenue is more than 1.1 times its 2018 revenue. The proposed rationale is to combine recent price movement,…
This Chinese-language post describes an A-share stock screen combining three conditions: a 14-period RSI below 65, an outer-to-inner trading volume ratio above 1.3, and circulating market value between 5 billion and 10 billion yuan. It outlines a screening…
This example turns a CAPM regression into a monthly stock-selection process. It takes a recent window of daily returns for eligible constituents, adjusts stock and benchmark returns by a stated daily risk-free rate, and regresses each stock’s returns against…
This article introduces the autoregressive moving-average model as a combination of AR terms, which use past observations, and MA terms, which represent past shocks. It describes choosing the orders p and q with autocorrelation and partial autocorrelation…
This Chinese stock-selection rule screens for companies associated with the metaverse theme, with the previous day’s actual turnover rate between 3% and 28%, while excluding stocks on the STAR Market. The stated rationale is that turnover may reflect…
These notes summarize ideas from a Chinese trading book through ten named principles and effects. They cover how payment frequency shapes perceived gains and losses, how unknown factors and nonlinear systems complicate market decisions, and how penalty kicks…
The article explains how WorldQuant’s 101 formulaic alphas combine short horizon price and volume features, often mixing momentum and mean reversion. It distinguishes signals traded on the same day as their latest input from those traded later, and walks…
This index timing method fits a quadratic function to a local segment of a historical price series, using either closing prices or the average of opening and closing prices. It treats the slope at the newest fitted point as an indicator of whether the series…
The document explains how the Capital Asset Pricing Model can be used to assess stock returns relative to market risk. Under CAPM, expected return is linked to the risk-free rate and the stock’s beta multiplied by the market risk premium. A regression of a…
This post describes a stock screen for main-board shares that combines a daily turnover-rate band of 3% to 12%, a circulating market value between 5 billion and 10 billion yuan, and an additional company-type criterion chosen by the user. It gives equivalent…
This stock-selection example filters Chinese equities by a turnover rate between 3% and 12%, a K indicator below 20, and a daily price change between -5% and 2.6%. The article presents the screen as a way to find stocks with potential, then suggests adding…
This tutorial explains how support vector machines classify data by finding a boundary with a wide margin, and how slack variables allow some classification errors in noisy data. It introduces kernel methods as a way to handle nonlinear boundaries by…
The post describes a stock screen requiring MACD above zero, a positive price-to-earnings ratio, and positive institutional direction. It treats the MACD filter as an upward-trend signal, positive earnings valuation as a basic financial condition, and the…
The document describes a short-term forex approach that opens a position around the transition between trading days, following the direction indicated by the previous day’s candle. It also discusses a script designed to collect statistics on whether a price…
This technical stock screen combines RSI below 65, three consecutive bearish candles, and a MACD reading above zero. The intended idea is to find stocks that have recently pulled back while the broader indicator remains in positive territory. The article…
This Chinese equity screen selects stocks with RSI below 65, free-float market value between 5 billion and 10 billion yuan, and displayed best-bid volume greater than best-ask volume. It combines a technical indicator, a company-size filter, and a snapshot…
This Chinese equity screening rule selects stocks whose previous day’s price amplitude exceeds 1%, that appeared on the prior day’s trading leaderboard, and that rank among the top five by current-day auction amount. The proposed rationale is that elevated…
This Chinese stock screen combines a turnover band of 3% to 12% with a circulating market value between 5 billion and 10 billion yuan. It then uses a comparison between the latest daily low and the previous day’s low as a short-term price filter. The…
This stock-selection screen targets companies associated with China’s metaverse theme. It filters for prior-day actual turnover between 3% and 28%, market capitalization below 10 billion yuan, and positive earnings per share. The document explains these…
This Chinese equity screen selects stocks with RSI below 65, a positive return over the prior ten days that remains below 35%, and no limit-up move on the previous day. The article presents these filters as a way to identify stocks with recent gains while…