/explore
|
You are here 
|
indrajeetpatil.github.io | ||
| | | | |
aosmith.rbind.io
|
|
| | | | | Where I discuss simulations, why I love them, and get started on a simulation series with a simple two-group linear model simulation. | |
| | | | |
freerangestats.info
|
|
| | | | | Stepwise variable selection is bad and dangerous, and you shouldn't do it. It increases false positives. It drops variables that should be in the model. It gives biased estimates for regression coefficients. The problems are worse for smaller samples; higher correlation between the X variables; and models with weaker explanatory power for the y (i.e. lower R-squared). | |
| | | | |
svmiller.com
|
|
| | | | | Here is a how-to on bootstrapping standard errors in R in a flexible way, using some tidyverse-friendly packages like modelr and purrr. | |
| | | | |
brenocon.com
|
|
| | | [AI summary] The provided text is a collection of comments and discussions from a blog post that originally criticized artificial neural networks (ANNs) in 2008. The comments reflect a range of opinions and debates about the relationship between machine learning (ML) and statistics, with some users defending ML techniques like support vector machines (SVMs), probabilistic graphical models, and deep learning, while others argue for the importance of statistical methods. There are also discussions about the need for better communication between disciplines, the limitations of ML approaches, and the importance of understanding the underlying assumptions of models. The text includes recommendations for textbooks and courses, such as 'All of Statistics' and Andre... | ||
