For sometime now, I have found writing down notes on a computer (uploaded to an appropriate host such as google docs) useful. I do this mainly for stuff I read on history, (non-mathematical) economics, (non-mathematical) finance and news related topics. However, I also read quite a bit of stuff that requires math, eg. journal papers in statistics and economics. I know how to use Latex but it's a little cumbersome for note-taking since I have to type out what I have already written by hand. Recently, I realized that I often forget important arguments that I have read (probably because I never fully understood the material?) and I want a more disciplined way to do things.

I was wondering if any of you do the same and whether you have any advice. I could:
1) write the material in LaTeX format and then upload this as a text file and also it's pdf output online (the pdf output is needed as I may need to access my notes on a computer without the compiler) OR
2) I can write out the material in Google docs itself (it has an equation editor)

The first option seems more inconvenient since I have to upload two documents. The second option seems more convenient because everything appears on the same document and I can access and edit the material anywhere. The disadvantage of the latter if I wish to turn the whole thing into a more substantive work (so a technical report ), I have to click on each equation separately and then copy the equation onto my Tex file.

I would like to hear your thoughts on how you organize your notes or thoughts. Thanks!

Hi! I'm here with another request for a reference.

I'm working with linear mixed modeling to analyze some tricky repeated measures data: Participants provided saliva samples three times each day, for three days, at three separate timepoints (baseline, 3 months, and 10 months). That is, we have samples nested within days nested within timepoints nested within participants.

I'm looking for advice on how, theoretically or in any piece of software, to specify a covariance structure for the repeated measures. The issue is that the model should predict different covariances depending on whether the saliva samples in question are within the same day, on different days, or at different timepoints.

( more details below the cut... )

"Every time I check the exponent, it gets worse. Maybe I should stop checking it."

- me

Courtesy of Abstruse Goose. Discuss.

A point X is chosen randomly from the interval [0,1]. A second point Y is then chosen randomly from the interval [X, 1].

Showing that E(Y|X) = (1/2)(X+1) and E(Y) = 3/4 is straightforward. What I'd like to do is calculate E(Y^2) using E(Y^2|X)..... but I'm not sure how to do so: Y^2 is not uniformly distributed on [X,1]

Any thoughts?

You have a coin, a pyramid, and a die. The coin has the numbers 1 and 2 on it, the pyramid has the numbers 1, 2, 3, 4 written on it, and the die has the usual 1 through 6 on it. An item is chosen at random and flipped/rolled. The coin has probability .5 of being chosen, the pyramid has probability .3, leaving the die .2 probability of being chosen. Let Y be the number showing after flipping/rolling.

Find E(Y) and Var(Y).



My thought: let X be a random variable so that X is 1 if the coin is chosen, 2 if the pyramid is chosen, and 3 if the die is chosen. Then E(Y|x) = 1.5 if x=1, 2.5 if x=2, and 3.5 if x=3. So then I should find E[E(Y|X)] ??

This question has me all tangled up. I'm not sure I really understand conditional expectation. Any help is appreciated, thanks!

From Boing Boing,

What the $*@&#! is the Nephroid of Freeth?

And just to whet your appetites...

"ore commonly called Freeth's Nephroid (which makes it sound less like a tentacled devourer of souls and more like a little boy's pet monster), it's actually a special plane curve--which is also not as weird and confusing as it sounds...

...Strophoids are curves, but they're also things that happen to curves. Plane curves that get their panties in a bunch, if you will."

... you accidentally misread "northern lights" as "noetherian lights" and start thinking about DCC.
... you squirt mayonnaise on your sandwich in the shape of an aleph.
... you post such exploits on an online math community.
... you add to this list.

I couldn't find an active Computer Science LJ and figured there were some CS students here =), so here goes:

I'm applying to computer science graduate schools, but I'm not sure what subfield (e.g., AI, Theory, Computer Architecture) to talk about in my research interests. Does anyone know how much it matters?

It's not that I don't know what I want to study -- I do. (Theoretical Computer Science, in particular, complexity theory.)

The problem is, I focused a lot more on machine learning and NLP as an undergrad (most of my undergrad research was in it, and 2 of my recs are from my supervisors), so I think my application looks a lot stronger to a Machine Learning person than to a theory person. (I do have a decent theory background -- I majored in math, took some advanced theory classes, and did research with a combinatorics professor [though it didn't go anywhere] -- but I don't think I'm currently as strong].

So should I say that I want to study Machine Learning (I'm stronger in it, but not really interested anymore) or Complexity Theory (less experienced, but it's what I want to study). How much does it really matter -- is it easy to switch once you're actually admitted? (Not sure how choosing an advisor really works...)

Hey everyone,

I am a student currently taking a stats class, and having trouble with test statistics. For some reason this stuff just does not compute in my brain. I was wondering if anyone knew off the top of their heads the equations needed for a couple problems I have to complete. I have a list of equations, but they are all for proportions and I have a feeling they are not the equations that I need (this is what I get for being out of class for a week due to swine flu, eeek!).


Problem 1:
A study found that the mean number of hours of TV watched per day was 4.09 for black (N=101, Standard Error = 0.3616) and 2.59 for white (N=724, Standard Error = 0.0859).
a. What type of test should you run?
b. Construct Hypotheses
c. Conduct a significance test using an alpha-level of 0.01 and interpret.
d. interpret your P value
e. Construct a confidence interval and interpret.
f. Interpret as a ratio.

Problem 2:
An experiment of responses for noise detection under 2 conditions used a sample of twelve 9-month old children. The study found a sample mean difference of 70.1 and a standard deviation of 49.4 for the difference.
a. What type of test should you run?
b. Construct Hypotheses
c. Conduct a significance test using an alpha-level of 0.01 and interpret.
d. interpret your P value
e. Construct a confidence interval and interpret.


If anyone could help me - it would be MUCH APPRECIATED!