Survival analysis for ordinary belongings
Still GoingHow long do your umbrellas really last?
List the things of one kind you have had: when each arrived, and when and how it went - or that it is still going. The page draws a life table, a survival curve and the median life, and reads the curve's shape: do they die young, go by luck of the draw, or wear out?
1 The register free
| # | Name (optional) | Arrived | Went | How it went | Remove |
|---|
2 The actuary's note metered
One run reads the curve in plain words, writes a life note for every item and suggests one change aimed at the commonest way they go. The numbers stay the page's own.
Pricing as soon as the session is ready...
The reply could not be read as a note. Here it is as it came back:
The page's check
4 Saved tables
How it works
Each item's life is the months from arriving to going; an item that arrived and went in the same month counts as half a month. Items still going - and ones given away in working order - have not ended, so they count only up to their last month: in survival analysis they are censored, and leaving them out would make everything look shorter-lived than it is.
The Kaplan-Meier curve steps down at each ending by the share of items still in use at that age that ended then. The median life is the age where it first reaches 50%. A Weibull shape fitted to the same data reads the bathtub curve: under 0.8 the risk falls with age (dies young), over 1.25 it rises (wears out), in between it holds steady (luck of the draw). With fewer than four endings the page says too few to tell. All of it is an estimate from a handful of items, not a law.