Explore >> Select a destination
|
You are here 
|
www.karsdorp.io | ||
| | | | |
bambinos.github.io
|
|
| | | | | ||
| | | | |
twiecki.io
|
|
| | | | | [AI summary] This blog post discusses hierarchical linear regression in PyMC3, highlighting its advantages over non-hierarchical Bayesian modeling. The author explores how hierarchical models can effectively handle multi-level data by leveraging the 'shrinkage-effect', which improves predictions by borrowing strength from related groups. Using the radon dataset, the post compares individual and hierarchical models, demonstrating that the hierarchical approach provides more accurate and robust estimates, especially in cases with limited data. The key takeaway is that hierarchical models balance individual and group-level insights, offering the best of both worlds in data analysis. | |
| | | | |
easystats.github.io
|
|
| | | | | [AI summary] The article explains credible intervals in Bayesian statistics, comparing them to frequentist confidence intervals, discussing their computation methods (HDI, ETI, SI), and their implications for interpreting statistical results. | |
| | | | |
pradyunsg.me
|
|
| | | My response to the discussion topic posed in Python Packaging Strategy Discussion Part 1 had become quite long, so I decided to move it to write a blog post instead. This post then started absorbing various draft posts I've had on this topic since this blog was started, morphing to include my broader thoughts on where we are today. Note: I've updated this to cover an aspect of the recent LWN article on the topic as well. | ||
