"Make Regression Great Again"

The new comic strip by PhD Comics is timely.

We can talk later about maximum likelihood estimation.

P.S. I stole the title from Won Chang's comment on the Facebook post by PhD Comics.

1982 Study by Institute for the Future

Wow, this 1982 study by Institute for the Future is almost dead-on correct. Just wow.

They did get the bit about the N-party system wrong, however.

(Tip of the cap to Jimmy Lin.)

Social Awkwardness is a Scarlet Letter

I wished I lived in a world in which social awkwardness didn't result in quite so prominent a scarlet letter. I was in G & D's (a local ice cream place) and overheard a rather demoralizing conversation: The people at the next table were talking about socially awkward people as if we were a lower form of life---apparently, we're less nice, feel less, etc. than other people. I know I can dismiss them as a bunch of idiots, but I felt like utter shit after listening to that and I do believe that most people look down on me to some extent because of my social awkwardness. (One of the reasons I feel bad when I hear that stuff is my belief that so many people believe it to at least some extent.)

Obviously, I have been able to find people who aren't that way, but one of these days I would like to believe that the number of such people isn't so small. I go out of my way to find people who I feel are egalitarian in this respect, and that's one of the reasons that I have a lot of difficulty finding friends. Simply, this is a very hard search problem. (Though, again, I have obviously found some of these special people!)

One can get better at social skills---but only to a point. (And indeed I am better now than I used to be.) In the same way, one can get better at doing contour integrals---but only to a point. In each case, there is some level of natural ability involved, and I wouldn't dream of looking down on people if they had trouble with contour integrals.

I was planning not to blog on another steroids scandal, but...

I told myself that I wouldn't blog about the whole A-Roid thing (sorry, but I can't help snickering about the new nickname), but former Major Leaguer Doug Glanville has made an excellent point in his New York Times opinion piece, and I think that is worth bringing up. Namely, the 2003 tests---i.e., the one that A-Rod failed---were agreed legally by the players as a litmus test to try to gauge just how rampant performance-enchancing-drug abuse was in the sport. It was the massive failure (over 100 players) on those tests that led to the testing system currently in place. One of the conditions for those tests was that all results were supposed to be anonymous and without that stipulation that testing would not have taken place. So, while A-Rod gets skewered for being caught (and don't think for a second that he's the last big name who will get similarly skewered---one shouldn't be surprised to see any name!), let's also remember that basic privacy rights associated with anonymous testing are also being breached here. Why is this issue getting lost?

(Tip of the Cap: I found out about Doug Glanville's article from Rob Neyer's blog.)