value composite
What it checks
Cheap is cheap if it shows up across multiple yardsticks, not just one. We rank each company on book/price, earnings/price, cashflow/price, and EBIT/enterprise-value, average those ranks, and prefer the cheapest names trending up.
Mechanism
Single-metric value (B/M alone, the original Fama-French HML) has been crowded for two decades. Asness-Frazzini 2013 showed that averaging several normalized value metrics (B/M, E/P, CF/P, EBIT/EV) recovers the lost spread. EV-based metrics (EBIT/EV) work better than equity-only metrics on asset-light firms. We z-score each metric over the ticker's own 12-quarter history and average — long when composite z is in the top quartile and trend confirms, short when bottom quartile and trend confirms.
Signal rule
Composite z of B/M + E/P (TTM) + CF/P (TTM) + EBIT/EV (TTM); long composite-z >= z_thresh + 60d uptrend, short -composite-z >= z_thresh + 60d downtrend.
Data dependencies
fundamentals_quarterlyQuarterly fundamentals (income, balance, cash-flow) from FMP + SEC.
daily_pricesAdjusted-close OHLCV for every US-listed ticker; primary price feed.
Expected edge
- Paper alpha
- ~3-5% ann. long-only; 7-10% L/S (1990-2012 in the source paper)
- Paper window
- 1990-2012
~3-5% ann. long-only premium 1990-2020; 7-10% L/S in the Asness-Frazzini 2013 backtest
Example tickers where this is likely to fire
Illustrative only — the signal fires based on the live data, not a fixed list.
Related families
qmjQualityQuality-Momentum (Asness-Frazzini-Pedersen 2019 QMJ): combine quality (gross profit / assets, ROE) with momentum — long when both are in their respective top buckets relative to their own history. The canonical cross-sectional implementation would rank across the universe; we approximate with self-relative quintiles, with fundamentals available-date lagged 45d to respect filing latency.
rd capitalized valueAccountingStandard book-to-market based value factors mis-classify R&D-heavy firms as "growth" because R&D is expensed (not capitalized) under GAAP. Peters-Taylor 2017 + Eisfeldt-Kim-Papanikolaou 2020 show that adding capitalized R&D back to book equity restores the value premium for tech/biotech and reverses the apparent "value-is-dead" finding of the 2010s. We compute an R&D-intensity metric per ticker and overweight it when it's accompanied by a price drawdown — the intangibles-implied value play.
buyback driftAccountingCompanies announcing open-market share repurchases earn ~3-4% abnormal return over the subsequent 12 months. Effect is strongest in "value-distressed" announcers (book-to-market in the top quartile, recent price drawdown). We approximate buyback announcements by detecting net share count contractions in quarterly fundamentals — ΔSharesOutstanding < 0 over 2 consecutive quarters is a strong proxy for an active buyback programme.
piotroski f scoreQualityPiotroski (2000, JAR) showed that a 9-signal accounting score distinguishes winners from losers within the high-B/M (value) universe. Long F>=7 / avoid F<=2 added ~7.5% annualized over 1976-1996 inside the value tertile, and ~23% in the highest B/M decile. Modern replications through 2021 confirm the score still differentiates forward returns even without the value gate; works especially well in uncertainty regimes when low-quality distressed names are marked down faster.
Explore value composite on alphactor.ai
See which tickers this family is currently firing on, with live signals and rankings.