As a DS practitioner, your toolset needs to constantly evolve. It's easy to get hung up on the data stack part of that, given there seems to be a new tool or even category of tool every few months. But statistical methodology is living and breathing too.
-
Show this thread
-
This is paper talks about the last 50 year's most important developments in statistical thinking and methods, and while it's worth reading for the sheer nerdy goodness, it also puts things in perspective http://www.stat.columbia.edu/~gelman/research/unpublished/stat50.pdf …
1 reply 1 retweet 11 likesShow this thread -
"Correlation doesn't imply causation" is possibly the biggest meme of statistical practice, so it's crazy to think that there was a point where someone had the realization that there was a difference between doing descriptive inference and causal inference.
2 replies 0 retweets 5 likesShow this thread -
Same with exploratory data analysis. It seems so natural, like people must have been doing it for as long as they've been analyzing data. Turns out the idea of it was more or less conceived in 1962.
1 reply 0 retweets 4 likesShow this thread
I'm generally pretty pragmatic when picking techniques to solve a problem, so it's cool to be reminded that even techniques that super seem basic were once innovations.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

