Resumos e ideias principais, escritos pelo agente de investigação da Stratmill, dos livros, artigos científicos, artigos e código consultados pelos nossos agentes de IA. Cada página inclui uma ligação para o original.
3,481 documentos
This brief BigQuant platform discussion asks how to implement complex methods for combining investment factors and whether suitable operators or code are available. The reply first asks what kind of combination the user has in mind, then says that modules…
This weekly research recap combines several China equity signals. A market-timing model evaluates broad indices through moving-average ordering and convergence, and describes the prevailing trend as weak with short-term signals leaning bearish. An industry…
This note compares DeepAlpha deep neural network and convolutional neural network stock selection experiments in a small-cap universe. The author describes a daily strategy that buys five stocks and sells them the next day, using 57 factors drawn from 19…
This Chinese-language forum post outlines an ETF backtest idea: rank funds by their return over five trading days, filter for those whose average trading value over the same period exceeds a stated threshold, then select the top five with equal weights. The…
The note describes quantitative investing as turning scattered theories and personal experience into a flexible trading system refined through programmed trading. It emphasizes that rules can reduce emotionally driven decisions and help keep actions…
This competition FAQ explains how submitted equity factors are reviewed and how model training should be separated from evaluation. It warns that training on the test dataset, or fitting a model on a portion of data and then using that same portion to…
This post describes an author’s deep-learning model built with market data and shared through open APIs, alongside links to simulation and backtest examples. It discusses stock selection from large-cap and mid-cap Chinese equity universes and says users can…
This post describes a research exercise using a supplied quantitative template to examine three short-horizon stock factors: five-day momentum, five-day reversal, and five-day volatility. The universe is described as stocks priced between zero and two yuan.…
This factor-replication note divides stocks into two behavioral groups. “Team” stocks are described as low in both volatility and turnover: after rising, they are expected to attract continuation expectations, become overbought, and then retreat. “Coin”…
This study constructs fixed-income fund-of-funds portfolios from products managed by one fund company, considering a setting where the available fund universe is limited. It separates funds into low-, medium-, and high-risk pools based on fund categories,…
This short BigQuant support exchange explains how to obtain historical fund data for use in a backtest. It recommends reading the fund bar data source with a start date and end date, with the backtest's current date supplying the endpoint. The example…
The document outlines high-frequency trading as a way to act on short-term intraday signals and accumulate small gains across trades. It groups strategies into trend-based, spread-based, and market-making approaches, and argues that their returns may…
The document is a short discussion of difficulties encountered by a small-cap strategy in 2024. Its author suggests that adding a value factor may partly address those problems and emphasizes drawdown management or stop-loss actions as an important part of…
This document introduces two ways to estimate the duration of public bond funds. The stated inputs are portfolio holdings disclosed in quarterly fund reports and the fund’s regularly published net asset value. The purpose is to give investors a way to assess…
This brief note introduces the Aroon indicator and names its inputs: high and low prices, a lookback length, and upper and lower bands. It gives formulas for the Up and Down components. Each component is based on how many periods have elapsed since the…
This article outlines a way to study stock selection during broad market declines. It proposes identifying historical selloff days, then examining whether stocks that rose to the daily limit despite the market weakness continue to show strength on the next…
This short Chinese-language discussion addresses why total market capitalization may be classified as a valuation factor in one platform's documentation and as a size factor in research literature. Its answer is that valuation and size factors have some…
This essay argues that chasing every rising sector and stock can undermine judgment and trading discipline. It recommends limiting a watchlist to a small selection of stocks to focus attention and reduce the impulse to pursue every apparent opportunity. It…
This advice article presents four habits for discretionary equity traders: record trade ideas and their reasoning before entry, review how subsequent price action compares with expectations, wait for stronger opportunities instead of trading constantly, and…
The document compares two fields in a Chinese stock market one-minute bar dataset that includes order-book information. It defines ask_volume1 as the total quantity currently displayed at the best ask, the first sell level in a Level 2 order-book snapshot.…
The document asks whether feature importance can be obtained from logistic regression, including Lasso-regularized models. Its reply contrasts tree-based importance measures with coefficient-based models and notes that a platform with scikit-learn installed…
This brief BigQuant forum post asks how to build one feature from another when the first feature is a moving-average comparison. The desired first feature assigns a positive value when the five-day average exceeds the ten-day average and a negative value…
The article proposes a short-horizon trading routine built around weekly profit targets and strict loss limits. Its central rule is to enter during the first trading day of the week, exit the full position once total portfolio gains reach 2.5%, and stop…
This paper examines whether deep neural networks can learn delta hedging strategies for options when the idealized assumptions behind Black–Scholes—such as continuous trading and no transaction costs—do not hold. It compares recurrent neural networks,…