Timeline for What is the parametrization of 'survreg' in the 'survival' R package?
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 16, 2020 at 19:33 | comment | added | DIVOC | Thanks. Your answer is enough for me to proceed with the task I need to to. | |
| Jul 16, 2020 at 19:26 | history | edited | psboonstra | CC BY-SA 4.0 | added 389 characters in body |
| Jul 16, 2020 at 19:20 | comment | added | psboonstra | I see now that my answer was incomplete. I have edited it to include how to calculate the hazard for a general error term $W$. With regard to your comment: generally the baseline hazard is called 'baseline' because it doesn't include any covariates, that is, it common to all observations. In contrast, your expression $h_0(t\exp(-x^\top \beta/\sigma))$ is an expression of covariates $x$, so I don't quite understand what you mean. Hopefully my edited answer clears up any confusion. | |
| Jul 16, 2020 at 19:15 | history | edited | psboonstra | CC BY-SA 4.0 | added section on calculating the hazard |
| Jul 16, 2020 at 18:27 | vote | accept | DIVOC | ||
| Jul 16, 2020 at 17:21 | comment | added | DIVOC | So, just to confirm, if I want to plot the hazard associated to individual $i$, I need to plot the hazard $h_0(t\exp(-x^T\beta/\sigma))\exp(-x^T\beta/\sigma)$? Where $h_0$ is the baseline hazard (not necessarily Weibull). | |
| Jul 16, 2020 at 16:45 | history | edited | psboonstra | CC BY-SA 4.0 | cleaned up language regarding AFT assumptions; need the tidyverse |
| Jul 16, 2020 at 16:39 | history | answered | psboonstra | CC BY-SA 4.0 |