alpha-zoo

To benchmark cross-sectional alpha factors, you use the alpha_bench tool to calculate IC and IR performance metrics over custom time periods and specific market universes. The skill evaluates factor libraries like Kakushadze 101, GTJA 191, and Qlib 158.

Cross-sectional alpha factors are quantitative signals predicting stock returns across a market universe at a specific time. You evaluate them to identify which factors from libraries like GTJA 191 deliver the best predictive performance for your investment strategy.

Yes, you can analyze custom user-supplied factor data. The built-in factor analysis engine evaluates your own factor datasets alongside prebuilt libraries to generate performance summary reports for your quantitative research workflows.

You run IC and IR analysis on the GTJA 191 library by using the alpha_bench tool. Specify the GTJA 191 factor zoo, select your target market universe such as CSI 300, and define the custom time period to execute the factor analysis and generate summary reports.

Executing factor analysis requires integration with internal registry and bench tools. These dependencies enable you to browse the factor libraries, validate the data, and run the IC and IR performance evaluations across diverse market universes.