NuclearPhynance Archive

I frankly don't care if someone knows about stoch calc if you don't have an intuitive understanding of the most basic probability concepts. E.g. I've heard very mathematical explanations of Feynman-Kac but when I ask people to explain what it does in very basic terms they often can't. Myself, I got my first job, as an exotics quant, without ever having picked up a book on stoch calc (in the time since then I have though Smiley ).



my 2 cents for the future content of quant type roles is that they will be a lot more optimization based than anything else. So knowing about stoch calc could come in handy but "cutting corners in the right way" will be more important.

"I've heard very mathematical explanations of Feynman-Kac but when I ask people to explain what it does in very basic terms they often can't."



When interviewing prospective pricing/actuarial analysts, I have always asked them to explain credibility theory and how it applies to ratemaking in 50 words or less, without using a formula or the word "credibility."  Amazing how that zings by so many candidates with exams under their belts ...



"if you start with some extremely complex, difficult to handle theory chances are small that you will arrive at something that can be used in practice some day."



In a similar vein, I have been known to laugh out loud at seminar speakers and vendors who describe complex neural networks or 411-factor pricing models for private passenger insurance risks.  This has probably not been good for my career ... but I fail to see how more than maybe a dozen factors have marginal contribution, and despise complexity for it's own sake.  Occam's razor in action ...