Timeline for Bayesian parameter estimation with proportion data
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 11, 2019 at 12:40 | history | bounty ended | Jens de Bruijn | ||
| Jul 8, 2019 at 11:31 | comment | added | Jens de Bruijn | Thanks! If I would implement the solution of placing the error term in the transformation, how would you transform observations 0 and 1? Since inverse $\Phi$ of 0 and 1 are $-\infty$ and $\infty$, the statistical package I use (PyMC3) returns errors. The most easy solution I can think of is setting the observations to near-zero and near-one (e.g., 0.00001 and 0.99999). | |
| Jul 8, 2019 at 3:21 | history | edited | Ben | CC BY-SA 4.0 |
deleted 9 characters in body
|
| Jul 8, 2019 at 0:56 | history | answered | Ben | CC BY-SA 4.0 |