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.
79,386 documents
The document outlines a daily stock screen that requires price range to exceed one percent of the prior close, a positive price-to-earnings ratio, and a rising 30-day moving average. It interprets these conditions as combining recent trading activity,…
This document describes a Chinese equity screen combining three conditions: membership in the metaverse industry, an arc-shaped price pattern, and a weekly close crossing above its 30-week moving average. It presents the moving-average cross as a way to…
The document explains how multi-factor models describe asset returns through factor exposures, factor returns, and asset-specific residuals. It presents Barra as a framework for estimating portfolio risk by combining factor covariance with specific risk, and…
This proposed daily Chinese stock screen combines three conditions: an intraday high-low range above 1% of the prior close, at least two limit-up events over the preceding 500 sessions, and positive large-order net inflow during the afternoon. The post…
This proposed Chinese stock screen focuses on companies classified in the metaverse theme. It requires the current close to be above the five-day moving average and applies a prior-session price-change filter intended to exclude stocks that closed at the…
This proposed Chinese equity screen selects non-ST stocks associated with beverage and alcohol imports or exports, with turnover between 3% and 12%. Before 10 a.m., it ranks candidates by limit-up activity and takes the first five. The post also shows…
This research summary examines whether trading behavior can serve as a proxy for speculative intensity in Chinese A-shares. It studies four measures: idiosyncratic volatility, idiosyncrasy, price delay, and size-adjusted turnover. The proposed intuition is…
This research summary examines whether trading behavior can serve as a proxy for speculative intensity in Chinese A-shares. It studies four measures: idiosyncratic volatility, idiosyncrasy, price delay, and size-adjusted turnover. The proposed intuition is…
This article outlines a simple screen for Chinese equities associated with the metaverse theme. It selects stocks whose actual turnover rate from two trading days earlier falls between 3% and 28%, while excluding stocks that hit the upper price limit on the…
The document describes how to build daily return data for level-two industries and use it in stock selection. It proposes joining stock industry classifications with daily returns and float market capitalizations, then grouping by industry and date. Each…
This short description introduces the Fisher Transform, an indicator developed by John Ehlers. It says the indicator tracks current market movement and that its plotted levels may serve as potential turning points. The stated directional reading is simple: a…
These release notes describe Hummingbot 1.19.0, focusing on a developing modular strategy framework and early dashboard tools for managing bots and backtests. The framework separates market data candles, controllers that choose actions, and executors that…
The post outlines a Chinese equity screening idea that combines at least five overlapping moving averages, company or industry characteristics, and a 15-minute MACD histogram whose negative bars are getting shorter. It interprets moving average overlap as…
This stock screen selects shares whose daily amplitude exceeds a threshold, whose price is above the five-day moving average, and whose prior session was not a limit-up day. The article frames a larger daily range as potentially useful for short-term trading…
The proposed screen seeks Chinese stocks with turnover between 3% and 12% and a positive gain below 35% over ten days, then looks for three technical indicators to cross upward at the same time. The accompanying formula describes rising 12-period and…
This indicator draws two colored filled rectangles on a selected bar to show the space between two Keltner channel values. Its configurable inputs include the smoothing method and length, two deviation ratios, the applied price, band colors, the bar used for…
This tutorial develops an earlier cryptocurrency spot hedging bot for trading price spreads between two exchanges. It adds optional spot margin mode switching for Binance, separate trigger thresholds for trades in each direction, chart lines and live spread…
This brief summary of a 2018 Chinese new-share market review reports that IPO issuance slowed while subscription winning rates remained stable. It also says that new-share subscription returns differed by investor category: A- and B-class investors…
This overview explains active learning as a way to reduce the cost of building supervised or semi-supervised models when expert labels are scarce. A model repeatedly identifies candidate examples for human review, incorporates the resulting labels through…
This document describes a Chinese stock selection rule combining daily price range, recent limit-up activity, and a long-term moving average. Stocks enter the candidate pool when their amplitude exceeds 1%, at least one limit-up occurred in the prior 25…
This brief note describes a non-averaged, three-line trading indicator whose arrays are displayed as separate lines. Its updated version removes the period input, leaving no user-set formation parameters. The signal guidance centers on a yellow line moving…
This stock-screening proposal combines three conditions: a 14-period RSI below 65, first-level buy volume greater than first-level sell volume, and a rising KDJ component. It presents these as filters for stocks with moderate momentum, buying pressure, and…
This forum post raises a factor-construction question about accessing older financial statement observations beyond a platform's precomputed factors. The example is operating revenue: the author understands the suffix-zero field to represent the latest…
This post describes a Chinese main-board stock screen requiring turnover between 3% and 12%, a daily gain above 1%, and a nonempty name for an outstanding convertible bond. It presents the bond-linked filter as a way to identify companies with…