Print Story PRO TIP: Don't lock the knees.
Diary
By gzt (Fri Feb 05, 2010 at 02:04:38 PM EST) gzt, sigma, indices, syncope (all tags)
Somebody fainted on the train today. Probably locked her knees. Somebody also fainted at my wedding, apparently. Same thing. People, apparently, learn in choir not to do that when standing. I'm sure there's a joke to be made about syncopation/syncope somewhere.


Remember, kids: being strong is useful.

My ears are still ringing.

Class. There was a quiz and for some reason I blanked on the thing. It was either recite a theorem or calculate the variance of something, and I was like, "Frig! Theorem? Variance? What's a variance?!" Fortunately, most people apparently did the same. It's hard to get started in the semester. TO DO: make a sheet that tells me all this probability and linear algebra stuff, make a sheet that has important theorems and formulas on it.

I'm getting a doughnut. I've been terribly hungry the last few days. I think I have been eating less, and that explains it. I've been training irregularly. I think 4 times in 13 days, which doesn't sound so bad, since it should be 5 or 6, but that spread makes me intuitively eat less. Still, I've managed to put 5# on my squat each time I make it to the gym, which is what you expect when you're still putting on strength (today is 150kg). If all goes according to plan, and that's a big if, I'll be at 160kg for three sets of five before Lent. Not impressive, but decent, and something I can probably maintain while eating less. I'll certainly make it to 155kg, though.

When in college, one of the things everybody had to do was take either 2 or 3 quarters of a humanities sequence, there were several, but people referred to it as "Hum", pronounced like the first syllable of humanities. People often read Hume in these classes. There's a pun or something to be made somewhere. Anyway, the classes weren't so bad, you'd write 4-8 page papers arguing things, they weren't serious enough yet that you had to read secondary sources. They were the sort of things you could bang out if you read the book, paid attention in class discussion, and were opinionated and passionate about whatever you were reading, and that's all they were really looking for. I only share this because somebody mentioned Hume somewhere. I like Hume.

It's unfortunate, I always thought it would be great to take literature classes. I liked reading, I liked class discussions, but I wasn't particularly interested in writing papers that were much more than what I mentioned in the previous paragraph.

What I am getting more interested in: economics or statistics, probably less statistics. Statistics is more toward the realm of math that I don't like so much, with all sorts of ugly capital sigmas and indices, but it's a necessary evil.

< I feel sorry for you furriners that don't have the best health care in the world | Let's see if I can get some updates in >
PRO TIP: Don't lock the knees. | 9 comments (9 topical, 0 hidden) | Trackback
Statistics by gpig (2.00 / 0) #1 Fri Feb 05, 2010 at 02:26:12 PM EST
My current employment is teaching me a lot about statistics. It's great, though frustrating when I go back into the real world and read something that tells me "study shows X is better than Y" and I want to ask (a) what was the sample size and (b) what significance test did you use, and (c) at what confidence level was it applied. I have a horrible feeling that there are a lot of cases where the answer to (b) and (c) would be "what's a significance test?"
---
(,   ,') -- eep
study shows by gzt (2.00 / 0) #3 Fri Feb 05, 2010 at 03:08:00 PM EST
The thing is, pretty much every "study" that gets cited is significant at a p<.05. It's the default and what just pops out of a lazy man's stat routine. The real money is looking at the study design, which only rarely actually has carryover to what, exactly, you're discussing. This stuff carries over into a lot of the more fuzzy sciences, too. One example is that whole "vaccines==autism" thing. Okay, that's not such a great example of this because there are 20 billion things wrong with that situation...

[ Parent ]
True by gpig (2.00 / 0) #5 Fri Feb 05, 2010 at 03:11:06 PM EST
Everybody loves 5%. Sample size is also important though .... and often science writers will report a 20-person medical trial the same way they report on a cohort study of 200,000.
---
(,   ,') -- eep
[ Parent ]
yes, you've got a point there. by gzt (2.00 / 0) #6 Fri Feb 05, 2010 at 03:18:49 PM EST
And then, of course, you get out of science into things about consulting, where they don't do studies, they slap together nonsense about how doing specific things may have saved money in a particular context, and then make a "model" tying results of some survey to those "dollars" in general. And then we have a general conclusion they sell in general: such and such saves you money, buy our services.

[ Parent ]
That link on being strong by lm (2.00 / 0) #2 Fri Feb 05, 2010 at 02:27:47 PM EST
We're Sorry
We seem to be experiencing some technical difficulties.
Please come back again soon.

I've never really liked Hume. But now that I'm studying his Treatise on Human Nature, he's growing on me. He states things straightforward instead of being circumspect like he tends to be in his later works. Consequently, I can actually understand his point.


There is no more degenerate kind of state than that in which the richest are supposed to be the best.
Cicero, The Republic
the gist was: by gzt (2.00 / 0) #4 Fri Feb 05, 2010 at 03:09:28 PM EST
man lifts car off little girl. little girl okay the next day. huzzah.

I like Hume a lot when he's being snarky.

[ Parent ]
That's a good day by lm (2.00 / 0) #9 Fri Feb 05, 2010 at 07:13:06 PM EST
There aren't many days when one is in the right place at the right time and prepared to be a hero.

There is no more degenerate kind of state than that in which the richest are supposed to be the best.
Cicero, The Republic
[ Parent ]
I always lock my knees by Herring (4.00 / 1) #7 Fri Feb 05, 2010 at 03:41:44 PM EST
I have lovely knees and don't want them stolen.

christ, we're all old now - StackyMcRacky
just keep 'em... by gzt (2.00 / 0) #8 Fri Feb 05, 2010 at 03:42:24 PM EST
...stapled together.

[ Parent ]
PRO TIP: Don't lock the knees. | 9 comments (9 topical, 0 hidden) | Trackback