Phase 5: Prospective Pre-Registration of the Soft-Fresh PEAD Composite in US Small-Cap Equities (Live-Weight Forward Test, April 2026 ~)

Read full text Summarise with VIRAT
Authors: Sungmin Hong
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This registration locks the parameters of a three-component earnings-momentum composite signal for prospective Out-Of-Sample (OOS) evaluation beginning April 2026. The signal combines Time-Series Standardised Unexpected Earnings (TS-SUE, weight 0.60), Sloan accruals quality (0.25), and announcement freshness decay (0.15), applied monthly to the CRSP small-cap universe (below NYSE 50th market-cap percentile, ≥$50M, 2003 - present). The prior evidence motivating this registration is documented across three completed studies. Phase 3 (OSF: osf.io/f7r6j) established a null result for TS-SUE in the S&P 500 large-cap universe over 2003-2024 (block-bootstrap p = 0.1484). Phase 4 applied the identical signal to the small-cap universe over the identical window and found a strongly significant result (BB p < 0.0001, ICIR = +0.353), with the IC difference between universes itself significant (BB p = 0.0244). Phase 5 extended the signal with accruals quality and freshness components, producing a 3/3 Holm-significant result (FWER ≤ 0.10) over the 2010-2024 OOS period. Contamination Disclosure: The composite weights (0.60 / 0,25 / 0.15) were selected after observing the full 2010-2024 OOS window and are therefore post-hoc. This registration converts the identical parameters into a prospective test. All performance from April 2026 onward is independent of the weight selection period. Performance prior to this registration date must not be treated as a blind holdout in any publication or investor document. The locked parameters, universe definition, rebalancing rule, execution cost methodology, compound stop rule, and checkpoint evaluation criteria are specified in full in the attached pre-registration document. Implementation scripts are held privately; the methodology is fully reproducible from the attached document alone.

Fields of Study & Subjects

Financial Markets and Investment Strategies Stock Market Forecasting Methods Financial Analysis and Corporate Governance

Publication details

Journal / source
Open Science Framework
Publication date
1 January 2026
Language
EN