Morsi portfolio: selected World Shock Lab ML comparison method

An owner-written portfolio case study describes comparing machine-learning and linear predictions around economic announcements with prespecified tests and explicit negative findings.

L2Research workflow Source Project-documented Outcome Author-reported

Evidence & verification

Source level

Project-documented
Grounded in primary project documentation, not necessarily an empirical evaluation.

Outcome verification

Author-reported
Outcomes are reported by the team that built the system. The entry is source-checked against the team's own publications; not independently reproduced.

Sources checked

2026-10-19 (UTC)

Source scope: Portfolio README and selected owner-written World Shock Lab case study. Described study findings remain author-reported, not independently reproduced.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (2)

Research applicability

A selected portfolio-owned economic ML comparison method, with an author-reported negative finding.

Getting started

Read the linked primary sources and the stated adoption requirements. Reuse suggestions are editorial and untested; no setup recipe or trial was executed.

Maintenance

Sources last checked on 2026-10-19. Future availability, maintenance, provider terms and compatibility were not verified.

License

Case-study text, data and sibling-software rights were not independently audited.

Workflow & human oversight

Scoped L2 for the selected World Shock Lab case study within the original portfolio. Researchers define timing, preregistered splits, metrics and a decision rule, then compare models, score held-out data and examine uncertainty. This is not a level for every portfolio project or an autonomous trading system.

Workflow level is not an evidence score; higher is not better evidence.

Approach

The case study defines shock timing and prespecified splits/metrics, refits on earlier data, selects candidates on validation data and scores held-out announcements once. It distinguishes causal identification, reduced-form associations, forecasts and hypothetical scenarios; the negative ML finding remains author-reported.

Limitations

Portfolio README and selected owner-written World Shock Lab static text only. The separate ATLAS page is context, not the classified method; sibling repositories are not silently substituted for the portfolio identity. No release, live forecast, raw data, figure, test or model was audited/executed. The author reports no evidence that ML adds value; this is not independently verified superiority. Causal labels depend on assumptions, with timing, proxy, weak-instrument, small held-out sample and prospective-scoring limits. Future forecasts are not observed performance. No investment advice or trading action is supplied. Static pages/HEAD are moving and non-atomic.

Reproducibility

Source text was checked only within this scope: Portfolio README and selected owner-written World Shock Lab case study. No software, dataset, experiment or result was independently reproduced.

Adoption requirements

Untested suggestion: examine the selected method, data timing, design, baselines and uncertainty before a separate research trial.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.