Recent content by Ma Xie Er

That's great advice, but if you were in my position, what would you do?

I'm fresh out of school, working for a department where there's some office politics. A co-worker, call this person worker X, is friends with the boss. Worker X is using logistic regression to build a model, and whenever I question anything, the boss tells me worker X is too busy to answer...

I'm going to answer my own question. No, the predicted probability is not a likelihood. The likelihood is the probability density function, as a function of the data. That is L(p|y) = f(y|p), for a fixed y. The likelihood is telling you how likely p is for a specific value of p, given the data...

I was trying to find an easy interpretation of the predicted probabilities of a logistic regression model, when one of my coworkers claimed that the logistic regression model is a likelihood. Now, I know that maximum likelihood estimation is used to estimate the parameters, but I didn't think...

Oops. You I forgot that case. For ##j=1,.,,[[n/2]]-1##, I still don't see why it's true.

I think ##e^{ix}-e^{-ix}= 2 cos(x)##. In this case, ##e^{4 \pi t j/n}+ e^{- 4 \pi t j/n} = 2 cos(4 \pi t j/n)##, so shouldn't it be ##\sum_{t=1}^n 2 (1 + cos(4 \pi t j/n)## ? And after this I'm still not sure how the series sums to 0.

Here's a link to the text http://www.stat.pitt.edu/stoffer/tsa3/tsa3.pdf. I was trying to solve Problem 2.10 on pg 77 (pg 87 of pdf). I don't quite understand footnote9, which is why I posted. I'm completely new to Fourier decomposition, so I'm having a hard time with this.

This text denotes i as √(-1)

No. n is a positive integer, and j= 1, ..., [[n/2]], where [[n/2]] is the floor or greatest integer function of n/2.

Yes, I can see that, for n=4, j=1, but it doesn't work for j=2, n=4. ∑t=14 e4π i t 2/4 =∑t=14 eπ i (2t) = (-1)2 + (-1)4 + (-1)6 + (-1)8 ≠ 0.

How are they zero on their own? If this is by De Moivre's theorem, then that doesn't apply to non-integers powers, i.e. (cos(x)+isin(x))n ≠(cos(nx) + i sin(nx)) for n ∉ℤ.

I'm reading "Time Series Analysis and Its Applications with R examples", 3rd edition, by Shumway and Stoffer, and I don't really understand a proof. This is not for homework, just my own edification. It goes like this: Σt=1n cos2(2πtj/n) = ¼ ∑t=1n (e2πitj/n - e2πitj/n)2 = ¼∑t=1ne4πtj/n + 1 + 1...