Factor construction · Quality screens · Systematic scoringTraditional quantitative value frameworks rank stocks by simple valuation ratios — price-to-earnings, price-to-book, enterprise value to EBITDA. These work on average but suffer from value traps: cheap companies that are cheap for good reason. Our approach augments valuation screens with quality signals derived from comprehensive financial data, filtering out companies whose low valuations reflect genuine fundamental deterioration.
Quality in our framework means consistency between what a company's financials show and what its communications convey. Companies where structured data reports strong metrics but qualitative signals reveal increased risk, aggressive accounting, or management uncertainty score low on our quality dimension. This cross-modal verification — structured data versus unstructured language — identifies the divergences that traditional quant models miss.
Rankings are updated systematically as new data arrives. Signal weights are determined by the interaction of value rank and quality rank: the deepest value with the highest quality receives the greatest analytical weight. Turnover constraints limit unnecessary recalculations, and sector-neutral scoring ensures the framework captures the value signal rather than sector patterns disguised as value.