Survival › Parametric fits

Weibull fit (uncensored)

A maximum-likelihood Weibull fit estimates the shape and scale parameters of uncensored event times.

What is Weibull fit (uncensored)?

Weibull is a 2-parameter generalisation of exponential. Shape k determines the hazard shape: k > 1 = increasing (aging / wear-out), k = 1 = constant (exponential), k < 1 = decreasing (early-failure-prone). Scale λ sets the time unit.

For *censored* survival data use parametric_surv with distribution = Weibull, which handles censoring properly via likelihood-based estimation.

When should I use Weibull fit (uncensored)?

Reliability data with no censoring (everyone observed to failure).
Engineering / quality lifetime studies.
When you specifically want shape + scale parameters rather than a Cox HR.

What data does it need?

One numeric column of event times (all observed, no censoring).

What does it report?

Shape k + scale λ + log-likelihood + Weibull-implied median = λ · (ln 2)^(1/k).

What does it assume?

Independent observations.
Weibull-distributed times.
No censoring.

How do I interpret the result?

Shape k tells you the hazard story: > 1 wear-out, = 1 memoryless (exponential), < 1 early-failure. Burn-in tests yield k < 1 in early operation, k > 1 later.