1. Data frequency. The engine can ingest annual, quarterly, or monthly data. At present, most series are sourced from the World Bank WDI on an annual basis, with scope to plug in fresher national or FRED/OECD feeds over time.
2. History window. For each indicator we work with a rolling history window (up to the last 40 observations). This keeps signals anchored in recent regimes rather than very old cycles.
3. Standardisation. We compute z-scores for the latest observation using (latest − mean) / standard deviation over the history window. This lets us talk about how unusual a level is in standard-deviation terms.
4. Signals. Qualitative labels such as “Strengthening above trend” or “Elevated but cooling” combine the z-score with the direction of change versus the prior observation.
5. Composite engines. The Growth, Inflation, Liquidity, and External engines are simple weighted combinations of indicator z-scores, rescaled to a 10–90 range for ease of interpretation.
6. Analogues. “Closest analogue periods” highlight historical points where the z-score profile is closest to today, as a quick pointer to prior regimes that looked similar.
7. Caveats. All data is subject to revision and publication lag. Some series may be missing or interpolated for certain countries. The dashboard is designed as a qualitative screening tool, not a replacement for detailed country and asset-level work.
This tool is for internal and educational use only and does not constitute investment advice or a recommendation to buy or sell any security or asset.