Author: Darren
Thoughts on AI as a statistician
(A follow up to this post. ) The development of AI tools for performing statistical analysis and training machine learning models is concerning to me. Will the tools hit a point where a user can simply state an analysis (like a prompt in a LLM), provide some data, and the AI will take over? Can…
Example of AI user error
I was sent a mathematical formula for the estimated glomerular filtration rate (eGFR) by a colleague (the medical/biological details and background of the formula are unnecessary). He copied the formula from an AI summary from a search engine (Google I think). A screen shot is below. The formula he sent is related to a formula…
A good #vss365 prompt
Back to sharing #vss365
Now on Bluesky.
Material possessions can be experiences
This paper made the news a few years ago, suggesting that purchasing experiences provides greater in the moment happiness over purchasing a material possession. I found papers on this topic going back many years. Here is one (the link to the pdf is here) from 2009 supporting this hypothesis. As with many things in human…
The cost of a free dog: vet bills
In the summer of 2005 I became the owner of black lab mix from a coworker. Or she took ownership of me. Either way the dog was free. Her name was Lucy and she was a great dog. A very good girl. The kind of dog that makes it hard to get another, since no…
An amazing book for improving your writing
I’m currently reading the book On Writing Well: The Classic Guide to Writing Nonfiction by William Zinsser. As someone who is remediating his writing after a lifetime of neglect it is an amazing book. Even if you just read the first few chapters you’ll learn much. To start: strive for simplicity. I am a wordy…
Which is the harder race: Tour de France or Vuelta a Espana
Now that the Tour de France is wrapping up, I found this paper on comparing the Tour de France and the Vuelta in terms of (how the authors define) difficulty. I like these types of studies. Here is one arguing the Dawn Wall in Yosemite is the hardest rock climb in the world. I need…
Is p-hacking acceptable for exploratory data analysis? (Part 1)
The problem of multiple comparison is well established in statistics. The multiple comparison problem occurs when many(!) hypothesis tests are performed on the same dataset. By doing so, the chance of a false positive result (e.g., finding a statistically significant result that is actually not) increases. For example, if performing 100 hypothesis tests each at…