Main content

Home

Menu

Loading wiki pages...

View
Wiki Version:
*Article:* <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;**The Bayesian analysis reporting guidelines (BARG) are explained in the *open access* article available at [https://www.nature.com/articles/s41562-021-01177-7][1]** *Example:* <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;**An example of applying the BARG is presented in the html document, [BARG-Supplement.html][2]**, within the Files folder named BARG-Supplement. **The HTML format is strongly recommended because of its dynamic components,** but PDF versions are available as [BARG-Supplement-CodeHidden.pdf][3] (in which computer code is hidden) and [BARG-Supplement-CodeShown.pdf][4] (in which computer code is visible). To view the HTML file, download it to your computer and then open the downloaded HTML file in a browser. <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;The example is conceptually simple but is very lengthy and painstaking in order to carefully illustrate all details of the BARG. *Reproducibility:* <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;A section near the end of [BARG-Supplement.html][2] explains in detail exactly how to reproduce the analysis in the example. *Must every item on the BARG be addressed in every analysis?* <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;In general, the BARG presents guidelines for aspects an analyst must at least carefully consider. Many points are essential (e.g., carefully explaining the model parameters, if using MCMC then presenting convergence and ESS for all parameters and derived values). Some points are specific to the BARG but important (e.g., if using Bayes factors then presenting minimum prior model probability for the posterior model probability to exceed a decisive posterior probability). Some points, however, must be considered but might also be deemphasized in appropriate circumstances. In particular, prior predictive checks might be superfluous in situations where it is obvious that the prior is very broad and has minimal influence. A prior sensitivity analysis might be superfluous when using obviously very broad priors and estimating continuous parameters (but prior sensitivity analyses are always appropriate when considering Bayes factors). A BARG-aware analyst should at least carefully consider all the points on the BARG and explain if any of them are not reported in depth. [1]: https://www.nature.com/articles/s41562-021-01177-7 [2]: https://osf.io/aj3dk/download [3]: https://osf.io/qms8k/download [4]: https://osf.io/3bdr8/download
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.