<t>Hi Guys, I got my PhD (in financial mathematics) in 2008, and I've worked for a major investment bank for 1 year, and a private investment firm for 2 years. However, I'm beginning to realize that my computing skills is always the bottleneck stopping me progressing forward (in the technological pe...

<t>QuoteOriginally posted by: eugenek2010To compensate a little for the kilo- thing, in the second example, you forgot a factor of g, and your muscles are nowhere near 100% efficient. (Probably closer to 20%.) Walking uphill increases your energy expenditure by an order of 1 kcal per meter of altitu...

Thanks for clarifying, now everything makes a lot more sense!

QuoteOriginally posted by: slopsI guess it is the common error - we speak about calories when we mean kilo calories.So the treadmill is displaying kilo-calories?

<t>Here's something even more crazy... I live about 1 km from work, but the walk involves about 200m of walking up a hill with an incline of 10 degrees. Hence, by walking to work, I am effectively lifting myself vertically upward by 200m*sin(10 degrees) = 34.73m. Multiply this by 95kg (my body weigh...

<t>Just did two sets of exercises in the gym after work:1. Jogging for 20 minutes, calories burnt = 200 (according to the calorie meter on the treadmill). 2. Weights - 50 kg weight with upward displacement of 70cm per move. It seems that, the gravitational potential on the 50kg gained each move = mg...

QuoteOriginally posted by: lesniewskiDo not do what frenchX proposes: the symmetrized KL divergence is not a good distance concept. Use .Any reasons in particular that you favor D(P_Model || P_Market) over D(P_Market || P_Model)?

<t>Thanks guys, it seems that the Jensen Shannon divergence is relatively simple to implement, I might give that a try. But I guess my other broader question is that, we know D(P || Q) is different to D(Q || P), so they must be measuring two different real world phenomenon. But what exactly are they...

<t>Hi GuysRight now I'm trying to find something that measures the difference between the statistical P measure (derived from a factor model), and the market implied measure Q. So the obvious candidate for that is the Kullback-Leibler divergence, except that it has the property D_. So given what I'm...

Thanks for the reply. I'll have a think about it. (4 hours later)Problem completely resolved. Thanks again.

<t>Hi EveryoneI asked Mathematica to evaluate the Fourier transform of $f(x) = e^{-x}$, and it returned $\hat f(t) = \delta (i+t)$, where $\delta$ is the Dirac delta function. I confirmed this calculation by taking the Fourier inversion of $\delta (i+t)$ (on Mathematica) and I did recover my origina...

<t>Thanks for everyone's replies, it has been quite helpful. I guess I want to ask one more question, this is especially directed to senior members: If you're hiring a quant, and you see an applicant who has a PhD, worked for an individual for 1 or 2 years, would you1. Think that the applicant can't...

<t>I will be graduating from my PhD at the end of this year, and so far I've got two job offers that I need to choose from. A. A quantitative analyst role from a major bank.B. Model building for a rich individual, the models he's interested in are statistical arbitrage types, and racecourse/sports g...

<t>Assuming you know how to simulate uncorrelated standard normals Z_1 , ... , Z_n. Then (using Einstein summation) the vectors U_i = D_ij Z_j , i=1,...,n, are normally distributed with covariance matrix D'D (where D' = D transpose). In fact, D is just the Cholesky decomposition of the covariance ma...

<t>If I'm entering a PhD program and my primary interest was in mathematical finance. I still want to leave the option open between working in the academia or a Wall St firm say after I graduate. Where would you think I should go out of NYU (maths dept) and MIT (maths dept)? Obviously, NYU is more r...