The hottest Substack posts of Of Aurochs and Angels

And their main takeaways
1 HN point 14 May 24
  1. In statistics, a collider can affect the relationship between variables, leading to false impressions. For example, controlling for a collider can change the true effect of one variable on another.
  2. Berkson's Paradox shows how relationships between variables can change when selecting samples based on certain traits, leading to counterintuitive results.
  3. The impact of weighting test scores in selection processes can alter the relationship between test scores and outcomes. Higher weighting can sometimes lead to negative correlations, despite the underlying positive relationship.
0 implied HN points 14 May 24
  1. The post discusses an analysis of Berkson's Paradox using code to investigate relationships and generate graphs.
  2. The code includes functions to run simulations, compute metrics, and plot graphs for the analysis.
  3. Results from the analysis are presented through visualizations showing betas, correlations, and the impact of data filtering on performance.