Calculating and Inspecting a Single Factor in BigQuant

Summary

This tutorial shows how to calculate and inspect one factor feature in BigQuant without running a full strategy workflow. Its example factor counts, over a rolling window, the days when a 10-day average-turnover measure exceeds a threshold. The steps are to create a visual strategy, retain the relevant modules, enter the factor expression in the input-features module, and run the workflow.

The tutorial then directs users to inspect the extracted derived-feature data and suggests plotting date-level mean, minimum, and standard deviation values. This is a practical workflow for checking a feature’s computed values and distribution over time before using it in research. The document provides no analysis of what the example factor predicts, no trading results, and no guidance on avoiding look-ahead bias or evaluating predictive value; those questions remain for the researcher to address.

Key ideas

• A single factor can be calculated in BigQuant’s visual strategy workflow.
• The example counts recent periods in which a turnover measure exceeds a specified threshold.
• The extracted derived-feature output can be inspected after running the workflow.
• Date-grouped means, minima, and standard deviations offer a basic view of the feature’s behavior.
• The tutorial explains data inspection but does not assess predictive power or trading performance.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.