Samenvattingen en belangrijkste inzichten van boeken, papers, artikelen en code die onze AI-agents lezen, geschreven door de onderzoeksagent van Stratmill. Elke pagina verwijst naar het origineel.
3,481 documenten
The document proposes a holdings-based labeling system for domestic active equity funds, addressing the risk that fund names and stated mandates can misrepresent actual exposures and that theme funds may drift across industries. It assigns each stock one…
This tutorial outlines how to construct financial factors from quarterly or annual statements, whose time intervals differ from daily price data. It recommends processing the raw financial series with Python rather than applying operators designed for…
This document explains a rolling Hurst exponent indicator intended to classify price behavior as persistent, anti-persistent, or close to a random walk. It estimates the exponent with rescaled-range analysis: price deviations are assessed over subsegments of…
This tutorial outlines a conventional trend-following strategy for Chinese stocks using short- and long-period moving averages. It describes entering when the short average is above the long average and exiting when it falls below, with trades placed at the…
The article discusses China’s stock-market trading disclosure known as the Dragon-Tiger List, which publishes information about large trades and trading seats during unusual stock moves. It argues that the data can help retail investors identify…
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…
This short learner reflection defines quantitative investing as using data to identify patterns that can inform investment decisions. It lists reduced need for continuous screen watching and a different way for programmers to examine markets as perceived…
This note reviews two studies on cross-sectional stock returns. The first asks whether a stock’s sensitivity to economic forecast uncertainty helps explain its returns. It reports that stocks in the lowest uncertainty-beta decile outperformed those in the…
This note defines abnormal turnover as a comparison between a stock’s recent average turnover and its longer-term average, using a 20-day window against a roughly 250-day window. It reports factor-test statistics and discusses whether the signal may be more…
This report illustrates the Black–Litterman portfolio framework with a hypothetical industry-allocation exercise. It assumes an investor forming a 2020 allocation at the end of 2019, then combines four subjective views with market-implied equilibrium returns…
This strategy selects China A-shares by circulating market capitalization, removes suspended and specially treated stocks, and holds the smallest-ranked names in equal weights. The described implementation targets 50 stocks and rebalances on the first…
A BigQuant user asks why data in an M4 workflow contains many repeated dates after missing-data handling, with different values on rows sharing a date. The reported test window is May 1 through July 1, while the displayed data begins earlier and has repeated…
The post presents a BigQuant SQL feature extraction query for Chinese stock data using the Resistance Support Relative Strength (RSRS) approach. It computes a regression slope of highs on lows, standardizes that slope using a rolling mean and standard…
This brief troubleshooting note addresses an error in a stock LSTM strategy template involving the training epoch parameter. It advises changing the model-fitting argument from the older `nb_epoch` name to `epochs`, while leaving the batch size and…
This brief article proposes analyzing the relationship between ESG scores and conventional style-factor exposures to understand how ESG investment exposure may be obtained. Its stated analysis concerns US equities and fund data. The central idea is that if…
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…
The author describes a China A-share swing strategy that first narrows the universe to selected high-quality companies and blue chips, then uses a machine-learning model to identify timing patterns. The training labels are based on relative performance…
This opinion article argues that abrupt political statements can create sharp market swings and gaps that quantitative firms may exploit through rapid information processing and order placement. It contrasts quieter periods, when it characterizes systematic…
This research summary considers adding investment themes to traditional multi-factor equity models. Themes are treated as possible sources of concentrated risk when they drive market moves, and as potential alpha sources when their returns carry information…
This report summary discusses how investment themes can be incorporated into equity multi-factor models as both risk exposures and potential sources of return. It recommends selecting themes whose constituents share fundamental drivers and show statistically…
This article explains a structure inference network (SIN) that augments object detection with scene context and relationships among candidate objects. It represents a scene with global image features, candidate regions as graph nodes, and pairwise…
This report summary reviews Chinese A-share market conditions for April 7–10, 2020. It compares industry performance strength and relative price-to-earnings valuations, naming the strongest and weakest groups. It also tracks market breadth through the shares…
This BigQuant forum fragment shows an initialization routine for a stock strategy. It sets the trading universe to the constituents of the CSI 300 and assigns an initial capital value through the strategy context. The snippet describes the default as one…
The document explains why Pandas may warn when code assigns values through a slice of a DataFrame. Such a slice may be a copy rather than a view, so an attempted update can fail to change the original data or leave the result unclear. The example uses a…