3D Kaplan-Meier Survival Analysis of Quantum-Tech Startups

Startup Cohort (3D)

Funding environment 1.0×

Clock speed 1.0×

Live Statistics

Months elapsed

0 / 60

At risk

Failed

Censored

Ŝ(t) now

1.000

Median survival

—

Ŝ(t) — Kaplan-Meier curve

x = months · y = Ŝ(t) — real product-limit estimate

How it works

Same real hazard-based process as the 2D version, rendered as a grid of towers instead of a chart. Each of the 60 startups is given a true failure month T before the clock starts, sampled from a Weibull hazard via inverse-transform sampling so the draw is genuine, not scripted:

h(t) = h₀ · m · k · t^(k-1),  k < 1  (decreasing "valley of death" hazard)
T = (−ln(U) / (h₀·m))^(1/k),  U ~ Uniform(0,1), m = funding multiplier

A tower's height tracks how long its startup has survived so far. While a startup is still at risk, its tower keeps growing every frame; the instant simulated time passes its drawn failure month T, the tower topples over and greys out — a real, individual event, not an animation cue. Any startup whose T lies beyond the 60-month horizon (or is still standing when you end the study early) is right-censored: it stops growing, stays upright, and is marked as a survivor rather than a failure.

The panel chart is the exact same live Kaplan–Meier product-limit estimator as the 2D sibling, recomputed from the towers' real fall times and censoring times:

Ŝ(t) = Π_{t_i ≤ t} (1 − d_i/n_i)

Orbit the camera to see the cohort from any angle — fallen (red) towers versus still-standing (green, greying to gray once censored) towers give an immediate visual read on cohort attrition that lines up exactly with the step-function curve.

Ŝ(t) = 1.000 · t = 0 mo

Tower — startup still at risk Toppled — startup failed Standing gray — censored (survived)

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.