Thursday, February 21, 2008

boobs: is there anything they can't do?

an email from my friend Asa, currently working on a grad degree from Columbia:

Below is a section of my R program for econometrics class. I got stuck for a bit and Colin Peer offered some help. Can anyone spot his contribution? If so, he will give you $5.

> print(cbind(bols,sebols),digits=3)
[1,] 0.871 0.0738
[2,] 1.903 0.0735
[3,] 3.007 0.0813
[4,] 4.063 0.0746

> confint (lm (y~X[,2]+X[,3]+X[,4]))
2.5 % 97.5 %
(Intercept) 0.7257972 1.016931
X[, 2] 1.7583147 2.048108
X[, 3] 2.8465362 3.167260
X[, 4] 3.9155509 4.209627

> boobs
Error: object "boobs" not found

> R2 <- 1-sum(e^2)/sum((y-mean(y))^2)
> R2
[1] 0.960695
> N
[1] 200

> R22 <- 1-(1-R2)*(200-1)/(200-4)
> R22
[1] 0.9600934

Mikolaj's comment:
Asa, are you going to fail your econometrics class if you don't find the boobs?

Colin responded:
hahaha. He was like, "Colin, do you know how to do a linear regression analysis with a p value of 2.5 or less on a recumbent threshold database?" I walked into his room and said, "Of course. That's why they made this program really, because it's so easy to do that on here with these sets." You should have seen his face light up. Like I just saved him a half hour of struggling with a new language. It made it that much better when I just wrote Boobs on the screen and walked away. hahahahahahahaha.


Ned said...

Sitting here in my office, on this beautiful snowy morning, looking out into the woods, it's nice to think about boobs.

Grant Miller said...

I run into similar problems when looking for boobs.