About

What this is

Lead/Lag Lab is a public research project testing whether alternative data signals — Wikipedia pageviews, GDELT news tone, Google search trends, and SEC filings — predict subsequent stock returns. The question is studied across the S&P 500 and a retail-attention basket, with rigorous methodology and honest reporting.

This is a portfolio piece and the seed of a real study. It exists to answer the question seriously, not to sell anything or generate leads.

Disclaimer

This is research only, not investment advice. Nothing on this site constitutes a recommendation to buy, sell, or hold any security. Past signal performance does not predict future results. Lead/Lag Lab is not a registered investment adviser. Do not make investment decisions based on anything you read here.

The prediction ledger is published for academic purposes. With the current sample size, directional accuracy is not statistically distinguishable from chance, and even when it becomes statistically significant, it does not imply economic significance after transaction costs, taxes, and execution risk.

Contact

Questions, errors, or data source concerns: [email protected]

Source code: github.com/leadlaglab-boop/leadlaglab

Data attributions

Full source documentation, terms, and what we store vs. publish:

  • Wikipedia Pageviews — Wikimedia REST API. Data: CC0. API docs
  • GDELT Project — Open data. Terms
  • SEC EDGAR — Public government data, no restriction. Developer info
  • Tiingo — Derived statistics only; raw price data not redistributed. Terms
  • Google Trends / pytrends — Unofficial, rate-limited, cached aggressively.
  • FRED / ALFRED — Federal Reserve Bank of St. Louis. Open license. Terms

Full details in docs/DATA_SOURCES.md.

Privacy

This site uses Cloudflare Web Analytics — no cookies, no personal data stored, no banner required. The watchlist feature stores your selection in browser localStorage only; no server ever receives it.

Security

Security policy and responsible disclosure: docs/SECURITY.md