APEX factor · Momentum

Momentum factor — explained

Momentum bets that stocks rising over the past 12 months (excluding the most recent month) keep rising for the next 1–6 months. The mechanism is behavioural — under-reaction to good news plus delayed institutional buying — not a fundamentals story.

Where this comes from

Academic anchor

Jegadeesh-Titman 1993 — Returns to Buying Winners and Selling Losers
Documents that ranking US stocks by trailing-12-month return (skipping the most recent month) and buying the top decile vs shorting the bottom decile earns ~1% per month over 1965-1989. The effect persists out-of-sample, in international markets, in commodities, and in fixed income. Asness-Moskowitz-Pedersen 2013 showed momentum and value work everywhere — not a US-equity quirk.
Plain English

What it actually measures

News diffuses slowly through the market. When a company beats earnings or wins a contract, the stock pops, but it takes weeks for the full institutional ownership chain to update — analysts revise estimates, index funds rebalance, momentum funds add the name. During that window the stock keeps drifting upward. Momentum factors capture that drift. The 'skip the most recent month' detail matters: very short-term price moves are dominated by mean-reversion (one-day reversals), so we exclude them. We measure 12 months ago up to 1 month ago.

No calibration constants

Math sketch

r_12_1 = ln(price_t-21 / price_t-252)         // 12-month minus most-recent month
m_sector = r_12_1 - mean(r_12_1 in same sector)  // sector-relative residual
momentum = z_score( w₁·r_12_1 + w₂·m_sector )

We z-score across the universe and across sector cohorts. Sector-relative residual matters because raw 12-1 momentum picks up sector beta (energy stocks all rise together when oil rallies), and we want the idiosyncratic portion that survives a sector rotation. The blend weights w₁, w₂ are not disclosed.

Pipeline

How DeepVane implements it

Daily price refresh runs every 5 minutes during market hours; the 12-1 return is computed off Yahoo Finance EOD data nightly. Sector membership comes from the universe table (GICS classification). The factor sits in apex_factor_scores keyed by ticker and date, and feeds the composite at scoring time. Cross-asset macro conditions (HYG/LQD ratio, DXY, VIX backwardation per the v11 add) influence whether momentum's amplifier is bumped or dampened that day.

One coherent posterior

How it composes with APEX

Momentum is regime-conditional in the strongest sense — its A_REGIME amplifier is +30% in risk-on and −40% in risk-off (Hamilton 1989 + Ang-Bekaert 2002). Combined with Value, momentum forms the AMP-V duo Asness-Moskowitz-Pedersen 2013 documented across asset classes. When momentum + quality + value align bullish on the same ticker, the GROWTH REGIME ALIGNED confluence pattern fires. Bearish momentum + falling earnings + high accruals triggers the QUALITY CRACK pattern.

Honest limitations

When it fails

Momentum collapses violently around regime turns. The classic failure is February-April 2009 — a 12-month winner cohort included financials that had crashed and then rebounded; the strategy bought them at the wrong time. The 2022 crash showed similar vulnerability. We hedge structurally via two layers: (a) BOCPD regime detector dampens momentum's amplifier when risk-off probability rises; (b) the conformal prediction interval widens automatically near regime boundaries so position sizing self-adjusts. Neither prevents a momentum crash; both reduce the bet size before one.

Read next

Related factors

QualityValue

See Momentum score on a real ticker

Every ticker page shows the per-factor decomposition. The Momentum score is one of twelve composing the 0–100 APEX composite.

Try NVDA →Full methodology35 invariants live