Timeline for How to fit a linear model in the Bayesian way in Mathematica?
Current License: CC BY-SA 4.0
23 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 19, 2021 at 6:20 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 146 characters in body |
| Jul 28, 2020 at 10:27 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | deleted 2 characters in body |
| Jul 28, 2020 at 9:29 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 1 character in body |
| Jul 28, 2020 at 9:23 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 12 characters in body |
| Jul 28, 2020 at 9:18 | vote | accept | Sjoerd Smit | ||
| Oct 2, 2019 at 15:44 | comment | added | Sjoerd Smit | @kglr I have never seen that before, but a cursory glance at that source suggest that there is no conjugate prior for that type of regression. Is that correct? It seems like a rather different method to me. At any rate, I just released an update to my GitHub code and included a function for doing Laplace approximations for inference problems. You could try to use that if you want a different error model than Gaussian errors. | |
| Oct 2, 2019 at 15:32 | comment | added | kglr | @SjoerdSmit, thank you!!! Any thoughts on how much additional effort it would take to extend it to quantile regression ( Bayesian Quantile Regression))? | |
| Oct 2, 2019 at 14:21 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 195 characters in body |
| Jul 29, 2019 at 8:55 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | Update link to point to a release tag instead (so future updates can't break the examples). |
| Jul 17, 2019 at 12:09 | history | edited | Henrik Schumacher | CC BY-SA 4.0 | deleted 6 characters in body |
| Jul 17, 2019 at 9:40 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 6 characters in body |
| Jul 17, 2019 at 3:12 | comment | added | BeanSith | Thank you for doing this work and sharing it with the community. I do find it frustrating that Wolfram has not made greater efforts to facilitate the use the Bayesian paradigm and statistical models for those users, such as myself, who are not capable of building/programming these models/implementations themselves; but nonetheless are able and desirous of following along with the competent and professional implementations of others. | |
| Jul 15, 2019 at 16:11 | comment | added | Eduardo Serna | Really cool stuff | |
| Jul 15, 2019 at 9:48 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 77 characters in body |
| Jul 14, 2019 at 17:18 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 1 character in body |
| Jul 14, 2019 at 15:49 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 357 characters in body |
| Jul 14, 2019 at 15:39 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 1953 characters in body |
| Jul 14, 2019 at 15:23 | comment | added | Sjoerd Smit | @chris Functions like LinearModelFit compute the Akaike Information Criterion and Bayesian Information Criterion. The serve a similar purpose, but are quite different from the Bayesian evidence. Furthermore, BayesianLinearRegression does not make point estimates (such as MAP) but retains the full distribution over the fit coefficients. Don't throw away information if you don't have to ;). | |
| Jul 14, 2019 at 15:16 | comment | added | chris | Nice. I believe version 12 provides penalty in all fitting routines of mathematica? This is equivalent to maximum a posteriori where exp(penalty) represents the prior? | |
| Jul 14, 2019 at 14:41 | history | edited | Sjoerd Smit | CC BY-SA 4.0 | added 77 characters in body |
| Jul 14, 2019 at 14:37 | comment | added | Sjoerd Smit | @Roman Yes you can: "Please refer to the example notebook for information about specification of priors" ;) (I updated that sentence with the section where you can find it) | |
| Jul 14, 2019 at 14:36 | comment | added | Roman | Awesome! Is there a way to set the prior? What prior did you use by default? | |
| Jul 14, 2019 at 14:34 | history | answered | Sjoerd Smit | CC BY-SA 4.0 |