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
This report discusses using optimization methods for asset allocation and portfolio construction, especially when objectives are complex, high dimensional, discontinuous, or heavily constrained. It argues that the in-sample optimum may not remain optimal in…
This Chinese-language document outlines how asset allocation models and fund portfolios can support wealth management. It frames portfolio design as a way to make investment choices more accessible to clients facing gaps in financial knowledge, behavioral…
This report extends a factor attribution framework for equity funds from quantitative strategies to actively managed funds. It argues that the factor set should reflect the fund’s likely investment approach and risk exposures. Active managers may share…
The article describes training BigQuant’s StockRanker machine-learning model using an overnight-return factor, following earlier factor research that links this signal to informed trading. Training uses all A-share stocks from 2013 through the start of 2021.…
This research summary examines how market-cap style relates to trader behavior and trading activity in Chinese equities. It argues that shifts between large- and small-cap stocks involve changes in capital flows, and that market capitalization is closely…
The report summary describes a strategy that varies fundamental factor exposures according to recurring seasons around company financial reports. It groups the calendar into four windows: annual-report disclosure, post-report performance trading,…
The document introduces an index-enhancement research report about building longer-horizon price and volume factors from mid- and high-frequency data. Its central motivation is a mismatch between data frequency and forecast horizon: as sampling frequency…
This overview explains how environmental, social, and governance considerations can enter investment decisions alongside conventional financial analysis. It surveys approaches including positive and negative screening, thematic allocation, impact investing,…
This brief strategy entry describes an A-share stock-selection example built around a k-nearest neighbors classifier. The accompanying discussion clarifies that the target is binary: a stock is labeled positive when its five-day return is above zero and…
This overview explains structured snowball options, focusing on how their coupons and risks arise. It describes coupon sources as exposure to the underlying asset's volatility, futures basis convergence for products linked to the China 500 index, and returns…
This research summary tests stock selection factors built from return and trading-volume data, using moments from the first through fourth orders. It places these measures in a broader taxonomy: factors can use one data dimension, combine dimensions at one…
The article argues that traders should avoid entering sectors after a short burst of broad attention and instead look for less crowded themes. It proposes treating three consecutive strong sessions and a sharp cumulative rise in leading stocks as a possible…
The discussion addresses how to combine daily observations with data from a longer period, such as the previous month, when constructing factors. Its suggested workflow is to convert daily market data into weekly or monthly bars with Python resampling, then…
This Chinese-language survey discusses fundamental allocation, mean-variance optimization, Kelly-CVaR, Black-Litterman, and risk-parity approaches. Fundamental methods include mapping economic-cycle phases to asset weights and using macroeconomic factors to…
This Chinese research abstract proposes using sales data from Pinduoduo to follow consumption patterns, particularly among consumers in China’s lower tier cities. It frames the analysis against rising household leverage, high housing costs, and slowing…
This brief discussion shows how to retrieve financial factor data through a data-access module using an SQL query. Its example selects dates, instrument identifiers, and a trailing-twelve-month operating-profit growth field from a Chinese stock…
This short strategy note describes an iterative experiment in which initial parameter adjustments appeared unpromising. The author then added a weight based on closing-price ranking and reports that the resulting Sharpe ratio exceeded 1.5. The takeaway is…
The document outlines a dynamic scenario alpha model (DCAM), which layers scenario classification over a conventional static alpha model. It argues that scenarios are useful when their subdivisions correspond to meaningfully different drivers or importance…
This research overview explains how knowledge graphs represent entities and their relationships, then outlines a construction process: gather data, preprocess it, identify entities and links, reconcile information from different sources, form relationship…
This research summary reviews a study of whether mutual fund managers’ personal investments in their own funds are related to fund risk. Using disclosures of investment ranges, fund holdings, and risk measures for actively managed U.S. equity funds from 2007…
This article summarizes research on how mutual funds’ trading decisions can lead other funds and how this herding relates to fund returns. It defines buy-leading and sell-leading funds according to whether peers follow their stock purchases or sales in…
This research note adapts the four phase investment clock to Chinese equities. It describes recovery, overheating, stagflation, and recession as distinct macroeconomic environments, and explains the traditional asset preferences associated with each. Because…
The document describes an integrated model for estimating how crowded equity factor strategies may be. It combines five standardized signals: valuation spreads between high and low factor-ranked stocks, short-interest spreads, within-group return…
The document describes an Expert Advisor that uses a single-layer neural network, identified as a perceptron, to classify the direction of quote movement. The implementation draws on the iAC indicator and bar opening-time history. It also reports a testing…