quant-factor-tracker

Track quantitative factor performance by computing Rank IC/IR and grouped long-short returns from cross-section factor exposures and subsequent returns, producing a standardized Markdown factor monitoring report with effectiveness summaries and failure alerts.

Rank IC decay analysis works by measuring IC mean, ICIR, and IC win-rate across multiple forward return horizons to assess predictive power stability, flagging factor failure or early-warning states when strong or weak conditions are detected.

Test factor monotonicity by splitting stocks into quantile groups to measure long-short performance, verifying that grouped returns exhibit monotonic behavior across the stock universe for the specified date range.

Yes, you need the gildata-aidata service strictly for data acquisition to compute ICIR and grouped returns, as the factor tracking workflow requires cross-section factor exposures and next-period returns sourced through this specific service.

Yes, you can assess factor effectiveness for a custom stock universe and date range by requesting the AI to generate a factor tracking report, which computes Rank IC/IR, grouped returns, monotonicity, and decay alerts based on your parameters.

Factor monotonicity testing may fail or show inconsistent grouped returns when the quantile long-short performance lacks monotonic behavior across the stock universe, triggering failure alerts or early-warning conditions in the standardized Markdown report.