知識圖書館

This forum exchange describes a fix for a BigQuant data query that failed with a missing date key while combining two data sources. The suggested query explicitly selects the date and instrument fields along with technical factor fields. It also filters…

This research summary forecasts constituent changes for China’s CSI 300 and CSI 500 indices ahead of their scheduled December 2018 reviews. It explains that the indices are reviewed in June and December, and that changes can create trading opportunities when…

This brief platform discussion explains how users can call MACD in BigQuant 3.0 after an older expression format stopped working. It gives two supported expression forms: a windowed MACD function and a shorthand helper. Both calculate MACD from closing…

A user reports a discrepancy between a trade’s recorded execution price and the stock’s historical prices on the same date after adjustment for corporate actions. The example contrasts a reported buy price with lower adjusted open and close values. The…

This excerpt summarizes a Chinese weekly research note on machine learning and commodity trading advisor strategies. It reports recent results for a neural network strategy on the CSI 500, a commodity futures strategy, and a commodity strategy combining…

This document explains how to remove ChiNext stocks from both the training and test portions of a visual AI stock-selection strategy on BigQuant. It instructs users to insert a shared “filter market” module before the missing-data handling module in each…

The report describes a multifactor stock-selection process that first evaluates and selects robust factors, then combines them into a composite score. It builds separate models for six stock universes and compares direct selection with portfolios that select…

The document explains how portfolio attribution compares a portfolio with its benchmark and decomposes excess return into drivers. It distinguishes return-based approaches, including Treynor-Mazuy, Henriksson-Merton, and related factor extensions, from…

This historical overview traces the independent development of the Capital Asset Pricing Model by Jack Treynor, William Sharpe, John Lintner, and Jan Mossin. It connects their different starting points to portfolio selection and corporate finance, and…

The study describes a machine-learning approach for classifying the optimal rebalance frequency of a two-asset portfolio strategy. It uses minute-level prices for 50 Binance crypto assets from 2022 and 2023, forms asset pairs, and groups them by correlation.…

This opinion piece argues that retail traders should be cautious about relying on remembered successes and discretionary short-term trading habits as markets become more influenced by institutions and quantitative models. Its central practical message is to…

This brief strategy outline proposes a weekly rotation among three of the smallest stocks that meet fundamental quality and growth screens. The listed criteria include sustained net-profit growth over three consecutive years and gross margins above 20%; it…

This presentation transcript surveys quantitative trading practices described for overseas markets. It explains how nontraditional inputs such as news sentiment, ESG reports, satellite imagery, and weather can be converted into structured signals, while…

This document introduces the Campisi model as a framework for explaining the performance of pure bond funds. It describes decomposing a fund’s total return and excess return into four sources: income, government-bond effects, spread effects, and security…

This article describes using recurrent neural networks to predict the direction of the next price change from high-frequency order-book data. The study uses data from Nasdaq stocks and keeps observations where prices change, reducing the input sequence. It…

This discussion raises a timing problem in backtests that use premarket processing. In the example, a signal is generated on one day and orders are submitted the next day; the author notes that historical data queried before trading may already include that…

This discussion describes a backtest problem caused by missing historical prices in a rolling maximum calculation. The author expected the portfolio to hold ten stocks, but in earlier periods it sometimes held only one or two. The suspected cause was that a…

This research examines whether industry rotation can enhance a CSI 500 benchmark portfolio. It compares two allocation approaches based on industry views: a moderate deviation from benchmark industry weights, and a concentrated approach that removes bearish…

This report evaluates a China ChiNext equity index that selects stocks using profitability, accounting strength, investment discipline, default risk, and low volatility. It argues that the index offers a more fundamental and lower-volatility alternative to…

This study compares two ways to build multifactor equity portfolios. Integration ranks securities by a combined score and selects those with strong exposure across the target factors. Mixing first builds separate single-factor portfolios, then allocates…

This analysis examines whether limits on how far bond yields can fall weaken bonds’ return contribution and diversification value. It models a global bond portfolio, a global 60/40 stock-bond portfolio, and a leveraged risk-balanced portfolio, using equal…

The document identifies the fitness-function setting used by a genetic programming module and lists its available objectives. These include information coefficient to information coefficient ratio, mutual information, long-only Sharpe, volatility and return…

This document summarizes research on how investment fund flows relate to fund performance and asset growth. It reports that international studies generally find a positive association between performance and inflows, with a nonlinear relationship to…

This forum post asks how to understand two approaches to adapting factors to market style. For a W-style split, the author proposes dividing a factor using a reference variable such as trading amount and treating the difference between the resulting groups…