Factor strategies, explained for non-quants

What a factor is

A factor is a measurable characteristic used to compare securities across a large universe. Quality, momentum and volatility are common examples in the academic literature. Historical relationships are not laws: a factor can lag for long periods, implementation costs matter, and no factor predicts one stock's next move.

Markov's implementation is its own rules-based research process. The papers linked below provide background for broad concepts; they do not validate a specific Markov strategy or recommendation.

Factor ideas in the public Markov methodology

Quality

Quality research compares measures such as profitability, balance-sheet strength and the persistence of operating results. Definitions vary. Markov's persistent-quality and quality-momentum lenses use documented company metrics to rank names relative to the research universe.

Momentum

Momentum compares recent relative performance. A raw momentum measure can also reflect a broad market or sector move, so it should not be read as proof that a company-specific development will continue.

Residual momentum

Residual momentum attempts to remove the portion of a stock's return explained by broad-market and sector exposure, then ranks the remainder. The method is a refinement of total-return momentum, not a guarantee that the residual will persist.

Low volatility

Low-volatility research compares realised price variability across securities. Combining a volatility screen with quality measures is intended to avoid treating low volatility alone as evidence of a strong business. A lower historical volatility estimate does not prevent a future loss or price gap.

Multi-factor composites

A composite combines ranks from more than one characteristic so that a single input does not determine the whole result. Markov uses the label QMJ for an internal quality-momentum-junk composite. That implementation is not identical to the academic “Quality Minus Junk” portfolio, although the academic work is useful background for the quality-versus-junk concept.

Why portfolios matter

Factor evidence is generally cross-sectional: researchers compare groups of securities, not a single “hero” stock. Diversification can reduce security- specific noise, but it cannot remove market risk or guarantee that a ranked basket will outperform.

The public methodology page explains how Markov applies these ideas. A company directory entry or factor label is not, by itself, a current recommendation.

Primary background sources

• Asness, Frazzini and Pedersen — Quality Minus Junk.
• Blitz, Huij and Martens — Residual Momentum.
• Fama and French — Common risk factors in the returns on stocks and bonds.

These papers report historical research, not expected Markov returns. Past performance is not indicative of future results.