The QVC Composite
Quality, Valuation, and a proprietary insider Catalyst — what survives an honest test
Abstract
We test a Quality–Valuation–Catalyst composite on the U.S. equity cross-section (2016–2026), pairing two classic low-turnover factors with DEHY’s proprietary insider-conviction signal as the Catalyst. Using point-in-time data, non-overlapping holding periods, and net-of-cost and sector-neutral checks, we report what survives rigor rather than what looks best. The durable, harvestable edge is a within-sector Valuation tilt; Quality’s headline strength is largely a sector bet; and the insider signal is a genuine freshness overlay rather than a standalone source of return.
01Introduction
A composite stock-selection model blends factors with low mutual correlation into a single signal. We follow the conventional Quality / Valuation / Catalyst spine, with one difference: our Catalyst is a proprietary, filing-derived insider-conviction score rather than a sell-side sentiment factor. This study asks a deliberately skeptical question — after costs, after sector effects, and out of sample, what part of the composite is real?
02Investment Philosophy
Our philosophy is quantamental: systematic factor construction disciplined by an honest reading of what the evidence supports. We prefer low-turnover factors with persistent return drivers, we validate each leg on its own before combining, and we publish limitations alongside results.
- Quality — Margins, return on assets, free-cash-flow margin, and low leverage.
- Valuation — Earnings, sales, and free-cash-flow yields — cheapness against price.
- Catalyst — DEHY’s insider-conviction score over recent purchases — the proprietary, freshness-sensitive leg.
03Methodology
- Universe & cadence — Every issuer with annual financials and price coverage (~3,100 names per quarter); quarterly rebalance, 2016–2026.
- Point-in-time — Financials are lagged by a reporting-availability window; forward returns are measured from the rebalance date. No look-ahead.
- Construction — Each leg is cross-sectionally z-scored; the composite is the equal-weight blend. We sort into quintiles and measure the top-minus-bottom spread.
- Metrics — Information Ratio, quarterly turnover, and returns net of a 25 bps round-trip cost — plus long-only and sector-neutral robustness checks.
04Results
As a long-short factor model the composite is strong and survives cost. But the long-short spread is concentrated in the short leg — distressed, low-quality names that are difficult to borrow and trade — so the decision-relevant question is how the long side behaves against a fair benchmark.
| Leg | Gross IR | Net IR | Quarterly turnover |
|---|---|---|---|
| Quality | 2.75 | 2.70 | 26% |
| Value | 1.14 | 1.11 | 29% |
| Catalyst (insider) | 0.10 | −0.01 | 21% |
| QVC blend | 2.56 | 2.46 | 44% |
Where the spread really lives
Every top-quintile basket underperformed a large-cap index over this window because the factor universe skews small- and mid-cap and the market was mega-cap-led. The long-short numbers are real but lean heavily on shorting distressed small-caps; we do not lead with them.
05The Decisive Test: Sector Neutralization
To separate genuine stock selection from sector bets, we demean each factor within its sector before ranking, then measure the size-screened top quintile against the equal-weight universe.
| Leg | Raw excess (IR) | Sector-neutral (IR) | Verdict |
|---|---|---|---|
| Value | +5.4% (0.81) | +4.9% (0.80) | Survives — genuine |
| QVC blend | +5.0% (0.76) | +4.0% (0.71) | Survives (rides Value) |
| Quality | +2.6% (0.92) | +0.0% (0.01) | Was a sector bet |
| Catalyst | −0.8% (−0.35) | −0.1% (−0.03) | Overlay only |
The honest conclusion
Within-sector Valuation is the durable, harvestable edge — roughly +4–5% per year over the equal-weight cross-section at an Information Ratio near 0.8, with low turnover and market-matching behavior versus the small-cap index. Quality’s apparent strength was largely a sector tilt. The insider Catalyst is a freshness and conviction overlay on a Quality-plus-Value base, not a standalone source of alpha.
06Strengths and Limitations
- Rigorous by construction — Point-in-time data, non-overlapping holding periods, net-of-cost accounting, and sector-neutral and long-only checks before any claim is made.
- Honest about the short leg — We report that the long-short spread depends on a hard-to-trade short book and present the long-only, size-screened result as the decision-relevant one.
- Limitation — regime length — A single 2016–2026 sample spans one broad regime; results should be read as indicative, not as a guarantee across all market environments.
- Limitation — capacity — The edge is strongest in small- and mid-caps, where capacity and liquidity constrain how much capital can be deployed.