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.
3,481 documents
The note describes two ways to address an ErrorReturnCode_84 encountered when running a BigQuant strategy with a StockRanker module. One option is to increase the AI Studio compute configuration. The other is to reduce the dataset and lower the model’s…
This research summary discusses how the balance between upward and downward stock volatility may help explain differences in future returns across stocks. It defines Relative Signed Jump Variation (RSJ) by taking the difference between upside and downside…
This overview introduces common stock-selection factors and gives formulas for valuation, momentum, quality, company size, and volatility. It describes price-to-earnings and price-to-book ratios, a twelve-month price comparison, return on equity, market…
This factor research note proposes a general method for designing trading factors by dividing a target variable into informative parts and recombining or transforming them. It identifies three elements: an additive object to split, a discriminating variable…
This tutorial introduces TensorFlow concepts, including tensors, computation graphs, automatic differentiation, and optimizers, then outlines a financial time-series example using a long short-term memory network. The workflow collects historical stock…
This overview explains how a stock-selection model can combine several signals to rank securities or construct a portfolio. It surveys market and systematic-risk measures, valuation ratios, growth and profitability metrics, momentum or reversal, trading…
The document outlines conservative investing as an approach that prioritizes limiting losses and seeking dependable returns over pursuing the highest possible growth. It presents long holding periods and restrained trading as ways to reduce the influence of…
This Chinese strategy note describes turning discretionary stock-screening rules into Boolean model features for a CNN classifier. The proposed universe favors companies with positive profitability and valuation measures, short-term strength, positive…
This research note forecasts constituent changes for five major Chinese equity indexes: the CSI 300, SSE 180, SSE 50, CSI 500, and ChiNext. It explains that index providers review constituents in June and December under published rules, and that large…
This brief presentation listing outlines a toolkit for high-frequency factor research. Its workflow covers fast calculation of historical factor values and storage, routine batch updates, adding factors to a research database, factor analysis, and rapid…
This introductory note presents Pandas as a library built on NumPy and describes its role in handling data in a quantitative research platform. It focuses on DataFrame objects, the two-dimensional tabular structure used for much of the platform’s data, and…
The document examines how the relationship between price-to-book valuation and expected return on equity can help assess expected earnings revision factors. A mathematical derivation motivates a linear relationship between log price-to-book and expected ROE;…
This research note describes a stock-selection factor built from the relationship between intraday changes in net buy orders and high-frequency returns. Net buy-order changes serve as a proxy for shifts in investor buying intent, while returns represent…
This Chinese A-share research summary examines airline stocks against the CSI 300 and proposes a fundamental timing and stock-selection model. It characterizes the sector as high beta and reports that relative performance has tended to vary by quarter, with…
The article discusses why leading Chinese quantitative private funds have become less visible among the largest shareholders of listed companies. It reports that several prominent firms and their products had appeared on such shareholder lists in earlier…
This report proposes using public mutual-fund holdings in Chinese equities to build portfolios in two stages. First, it selects funds by Sharpe ratio and rebalances the selected group quarterly. Next, within those funds' ten largest holdings, it identifies…
This Chinese-language report summary surveys how artificial intelligence may be applied across asset management. It describes hedge funds and investment banks adopting AI, then discusses potential uses in investment research, trading, risk management,…
This quarterly review examines China's equity index fund market in the second quarter of 2018. It describes faster ETF and LOF product development across broad-market and sector indexes, alongside the shrinking assets of leveraged graded funds. The report…
The article examines whether alternative equity weighting strategies outperform market capitalization weighting because of factor tilts or because they capture diversification return. It defines diversification return as the difference between a portfolio’s…
The article presents four ways to inspect a random forest: permutation-based feature importance, variation among individual tree predictions, per-observation prediction paths, and partial dependence plots (PDPs). For permutation importance, it compares a…
This research summary examines what happens when a brokerage resumes coverage of a stock after at least six months without coverage. It reports that the market response is especially pronounced when the original analyst returns. The response to a resumed…
This research review examines why the traditional book-to-price value factor underperformed growth stocks over an extended period. It argues that book value misses internally developed intangibles such as software, brands, research, and customer…
The document outlines Two Sigma’s systematic investment approach and briefly describes its China macro strategy, which invests across Chinese commodity and financial futures. It presents a process built around research, portfolio construction, and risk…
This study explores genetic programming as a way to discover formula-based signals for commodity futures trading. It adapts a genetic programming package to search combinations of market data and operators, aiming to find signals that could complement…