- July 19th, 2004, 11:27 am
- Forum: General Forum
- Topic: Lack of Citation linkage
- Replies:
**21** - Views:
**184162**

<t>QuoteOriginally posted by: caroemattcushman: "As far as anyone can tell, economics and physics have no link except via mathematics...."Leon Walras and Vilfredo Pareto were heavily influenced by physics when they started out their great project of what was eventually to become microeconomics. Insp...

- July 19th, 2004, 11:21 am
- Forum: General Forum
- Topic: Lack of Citation linkage
- Replies:
**21** - Views:
**184162**

<t>QuoteOriginally posted by: Collectormattcushman: "As far as anyone can tell, economics and physics have no link except via mathematics...."I would not be so sure on that Care to give an example? Well, I will agree that economics is a generalization of phyiscs... albiet very general. physics -> ch...

- July 18th, 2004, 10:58 pm
- Forum: General Forum
- Topic: Lack of Citation linkage
- Replies:
**21** - Views:
**184162**

<t>QuoteOriginally posted by: AaronWhile arrogance can spring from insecurity, it can also spring from talent. A few arrogant people have something to be arrogant about.That's true, and I'm sure it holds for some people in this case, but not all. Just like plenty of physicists who do work in finance...

- July 17th, 2004, 11:30 pm
- Forum: General Forum
- Topic: Lack of Citation linkage
- Replies:
**21** - Views:
**184162**

<t>Math and science types that willfully disregard the literature in a field that they are working are arrogant... though the arrogance is often actually a sign of insecurity. They are afraid that a deeper understanding of the field will force them to come to terms with the reality that their result...

- July 12th, 2004, 8:45 pm
- Forum: Brainteaser Forum
- Topic: "Equation"
- Replies:
**37** - Views:
**185771**

Similar question:solve the following "ODE":f = sum_k=1^\infty f^(k)where f^(k) is the k-th derivative of f.

- July 12th, 2004, 7:56 pm
- Forum: Brainteaser Forum
- Topic: Candy factory
- Replies:
**13** - Views:
**183528**

<t>I would say 105 choose 5, i.e. 105!/(100! 5!). This is considerably smaller than 6^100.... This is a fairly standard combinatorics question. You can see the mapping between the colorings and the set of subsets of 0..104 in the following way. Let {a1,a2,a3,a4,a5} be the subset. Then color a1 of th...

- June 29th, 2004, 8:41 pm
- Forum: Brainteaser Forum
- Topic: Prime Numbers
- Replies:
**9** - Views:
**187921**

I believe it's approximately 8,000,000,000,000,000/ln(8,000,000,000,000,000). People much smarter than I with more time (and frankly, interest) to figure this out have worked on this problem extensively.

- June 29th, 2004, 4:23 pm
- Forum: Brainteaser Forum
- Topic: Rock, Paper, Scissors.
- Replies:
**19** - Views:
**191603**

<t>It's not that tricky... I think you will immediately see that the probability of the next coin flip being a head is >> 0.5 conditioned on the previous one being a tail. Of course, you can do better with a longer history of the flips even.Being the brainteaser forum, here is one: calculate the pro...

- June 29th, 2004, 2:44 pm
- Forum: Brainteaser Forum
- Topic: Rock, Paper, Scissors.
- Replies:
**19** - Views:
**191603**

Exactly. Even if you take a uniformly random sequence of letters from the alphabet there will be a pattern.Of course, you can just take consecutive pairs of words.

- June 28th, 2004, 6:30 pm
- Forum: Brainteaser Forum
- Topic: Rock, Paper, Scissors.
- Replies:
**19** - Views:
**191603**

<t>QuoteFor an easy binary random number generator, I use quotes. I just pick a phrase I remember and pick heads if the next word comes earlier in alphabetical order, tails otherwise.Aaron, I'd love to play you in this game for money. This strategy implies a correlation between your coin flips.... <...

- June 24th, 2004, 7:19 pm
- Forum: Brainteaser Forum
- Topic: Prime Numbers
- Replies:
**9** - Views:
**187921**

<t>Let f(k) be the number of primes between k and k+999. We know f(1) is larger than 5, and there exists k_0 with f(k_0) < 5 (e.g. use kr's construction).Note that f(k+1) is either equal to f(k)-1, f(k), or f(k)+1 depending on whether k and/or k+1000 are prime.So, to get from f(1) >5 to f(k_0) < 5, ...

- June 14th, 2004, 2:27 pm
- Forum: Technical Forum
- Topic: Estimating An Infinite integral
- Replies:
**24** - Views:
**192155**

<t>Etuka: oh yeah, you're right, the last "trivial" step with the differentiating is more complicated that I thought. I didn't read the other threads carefully enough.kr: I don't think that there are any convergence difficulties... the only slight problem is that you are evaluating a Fourier transfo...

- June 14th, 2004, 4:19 am
- Forum: Technical Forum
- Topic: Estimating An Infinite integral
- Replies:
**24** - Views:
**192155**

<t>Etuka, I still think that my solution works and gives you something that should be explicit enough for most any purpose (not sure, but I think that spacemonkey's is somehow equivalent).To simplify, your I(x,n) is simply sqrt(pi/2) times the Fourier transform of e^(y^2/x^2) (1+y^2/A)^n, evaluated ...

- June 8th, 2004, 1:59 pm
- Forum: Technical Forum
- Topic: Estimating An Infinite integral
- Replies:
**24** - Views:
**192155**

In text, my equation says the integral from -infinity to +infinity of e^{-y^2/x^2} (1+iy)^n(1-iy)^n e^{2iy}. Hope this works out...

- June 8th, 2004, 1:28 pm
- Forum: Technical Forum
- Topic: Estimating An Infinite integral
- Replies:
**24** - Views:
**192155**

<t>I think you can solve this explicitly using Fourier analysis. The integrand is 1/2 times the real part of (that equaltion is missing a sqrt(A) in the denominator, but I don't think it matters too much.. I'm dropping factors all over the place here)This is the Fourier transform (w.r.t. y) of ((1-i...

GZIP: On