Cross Validated
2026-08-13 02:10 UTC
By adamkostanov
AI-113-20260813-social-media-4ba5344a
Do the shape and scale parameters have statistical interpertations?
Here is a Weibull PH model: $$h(t \mid \mathbf{x}) = h_0(t)\,\exp(\boldsymbol{\beta}^\top \mathbf{x}) = \gamma \lambda\, t^{\gamma - 1} \exp(\boldsymbol{\beta}^\top \mathbf{x})$$ Is statistical inference ever performed on the shape and scale parameters ( $\gamma, \lambda $ )? For example, do researchers ever comment on whether the estimates of the shape and scale parameters are statistically significant? Or what about determining the influence an extra unit change in shape or scale would have on hazard and survival rates (marginal effects)?
Here is a Weibull PH model: $$h(t \mid \mathbf{x}) = h_0(t)\,\exp(\boldsymbol{\beta}^\top \mathbf{x}) = \gamma \lambda\, t^{\gamma - 1} \exp(\boldsymbol{\beta}^\top \mathbf{x})$$ Is statistical inference ever performed on the shape and scale parameters ( $\gamma, \lambda $ )? For example, do researchers ever comment on whether the estimates of the shape and scale parameters are statistically significant? Or what about determining the influence an extra unit change in shape or scale would have on hazard and survival rates (marginal effects)?
Full article content could not be extracted automatically. Read the original below.
Source:
Cross Validated
· stats.stackexchange.com