Main Reading
The original rarity stack behaves like a tail-risk story: most people sit near common names, while rare surname/name/date combinations live far out in the distribution.
High Kurtosis (Fat Tails) suggests higher risk of extreme, rare events. Low Kurtosis (Thin Tails) suggests data is less prone to outliers, often indicating a flatter peak or uniformly distributed data. This page keeps the original Blair Page visual behavior while reframing the name-rarity, exact-date, and wealth-anchor information as a tail-risk dashboard.
The original rarity stack behaves like a tail-risk story: most people sit near common names, while rare surname/name/date combinations live far out in the distribution.
Blair + Burness + Drake/McCoy + exact date is modeled as an extreme outlier layer rather than an ordinary center-of-data case.
A flatter or uniform comparison has fewer extreme surprises and less tail concentration.
Browser fetch attempts World Bank population data and keeps safe fallbacks if a public endpoint fails.
How to read kurtosis against the original data idea.
Distribution shape, tail weight, component outliers, and live no-key API enrichment.
Same center, different tail behavior. Fat tails keep more probability far from the average.
Illustrative excess kurtosis values: uniform is low, normal is baseline, fat-tail models are high.
Original name/wealth components reinterpreted as tail intensity.
How much more tail attention each interpretation deserves.
Converted from “1 in N” to log10(N), so extreme profiles remain readable.
Keywords for framing the page.
Each card keeps the prior glass-card behavior but explains the statistical meaning.
Best single-page scan for high-tail and low-tail interpretations.
| Scenario | Kurtosis meaning | Tail risk | Practical interpretation |
|---|
High kurtosis means the dataset has more extreme observations than a normal-looking distribution would lead you to expect. In this page, rare name stacks and wealth anchors are treated as tail-event examples.
Low kurtosis means extreme outliers are less common. A uniform comparison can look flatter overall, with fewer dramatic spikes in the tails.
Kurtosis looks at the fourth standardized moment, so far-away values get amplified strongly.
Browser-only optional enrichment. The page works on GitHub Pages with static fallbacks even if public endpoints block the request.
World Bank population endpoint and static model values.
Public GET calls and local fallbacks.
Static links embedded for source review and later expansion.
This is a statistical visualization page, not a private-record lookup. The charts use illustrative distributions and the same public-style rarity/wealth anchors from the original page to explain how high kurtosis maps to rare outlier events. Use it as an infographic and teaching page rather than a formal statistical proof.