fats_tromino

5 ms·

1. 1.▲
   Benchmarks for Databases(github.com)

   1 points by fats_tromino 9y ago|0 comments
2. 2.▲
   by fats_tromino 9y ago

   I agree with this as well. I tried using Duckduckgo around a year and a half ago or two years ago and abandoned it due to poor search results, but I began using it again a couple months ago and it seemed much improved.
3. 3.▲
   by fats_tromino 9y ago

   This is the most succinct and clearest explanation of what's going on. I see this discussed a lot when people talk about the curse of dimensionality. Another very simple example is the example of a n-hypercube with edge length 1/2
4. 4.▲
   by fats_tromino 11y ago

   Thanks for the clarification about Mallow's C_p, I always get the name messed up for some reason.
5. 5.▲
   by fats_tromino 11y ago

   Thank you, I'm confused as to why this is happening.
6. 6.▲
   by fats_tromino 11y ago

   Also some features of regression and hypothesis testing are actually fairly robust to non-normality, particularly with large sample sizes.
7. 7.▲
   by fats_tromino 11y ago

   Look at normal qq-plot to measure normality
8. 8.▲
   by fats_tromino 11y ago

   Just as two quick comments, confidence intervals != prediction intervals. One gives a range around a true mean value, the other gives a range around where the variable may actually fall (prediction intervals are bigger). You may also want t