wjakethompson / measr

R package for the Bayesian estimation of diagnostic classification models using Stan
https://measr.info
GNU General Public License v3.0
8 stars 3 forks source link

Model Comparison #8

Closed wjakethompson closed 1 year ago

wjakethompson commented 1 year ago

Add functionality for calculating LOO and WAIC for estimated MCMC models. Also functionality for comparing model criteria for multiple models.

codecov[bot] commented 1 year ago

Codecov Report

Patch coverage: 95.16% and project coverage change: +4.57 :tada:

Comparison is base (76064f3) 89.04% compared to head (d9aa71d) 93.62%.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #8 +/- ## ========================================== + Coverage 89.04% 93.62% +4.57% ========================================== Files 12 20 +8 Lines 694 1381 +687 ========================================== + Hits 618 1293 +675 - Misses 76 88 +12 ``` | [Impacted Files](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson) | Coverage Δ | | |---|---|---| | [R/measrfit-class.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9tZWFzcmZpdC1jbGFzcy5S) | `100.00% <ø> (ø)` | | | [R/priors.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9wcmlvcnMuUg==) | `100.00% <ø> (ø)` | | | [R/stan-utils.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9zdGFuLXV0aWxzLlI=) | `88.00% <75.86%> (+13.92%)` | :arrow_up: | | [R/fit-dcm.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9maXQtZGNtLlI=) | `90.16% <77.77%> (+2.10%)` | :arrow_up: | | [R/data-checks.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9kYXRhLWNoZWNrcy5S) | `94.91% <80.00%> (-5.09%)` | :arrow_down: | | [R/utils.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui91dGlscy5S) | `92.30% <85.71%> (-7.70%)` | :arrow_down: | | [R/utils-methods.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui91dGlscy1tZXRob2RzLlI=) | `87.65% <87.65%> (ø)` | | | [R/utils-fit.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui91dGlscy1maXQuUg==) | `44.11% <93.33%> (+13.92%)` | :arrow_up: | | [R/add-criterion.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui9hZGQtY3JpdGVyaW9uLlI=) | `95.83% <95.83%> (ø)` | | | [R/utils-loo.R](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson#diff-Ui91dGlscy1sb28uUg==) | `96.42% <96.42%> (ø)` | | | ... and [10 more](https://codecov.io/gh/wjakethompson/measr/pull/8?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson) | | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=Jake+Thompson)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.