Biologist Quants

[LostBiologist]

Greetings all, and my apologies if the next few questions sound utterly clueless (I have searched for most of these on the forums, but I might have missed something). I'm currently half way through a PhD in biology in London, looking at biological noise in transcriptional networks. I'm becoming increasingly interested in quantative finance, and am currently working through Mark Joshi's book and improving my stochastic calculus. I'm wondering if some of you could answer a few questions.



Is a Biology PhD going to be a big disadvantage compared to maths/physics etc?



Are there many flexible internships available for prospective quants? I'm be very keen on doing an internship, but the PhD is going to be pretty hard to put on hold, particularly in my final year. Ideally I'd want something that I could do in parallel with my studies, perhaps one day a week or something (again, my apologies if that's a ridiculous idea).



Alternatively, is there a good qualification I could do in my spare time to improve my chances of getting a quant position? Someone I was talking to suggested the IMC, but I don't think that's right for a quant?



How is the jobs market at the moment? I'm kind of aware that the last few years have been pretty poor, but are people starting to hire again?

[lys]

I know a couple of people that used to do computational biology that made the leap. That was a while ago though.



Today it's a different reality and it's still very tough to get any job, especially entry level.



In any case with that background I would argue that learning option pricing and stochastic calculus is a bit of a waste of time. You are more interesting to the buy-side (hedge funds, prop-shops). Focus on signal processing and filtering, isn't that what you do?



I wonder why you want to switch? Computation biology/biotech is probably the hottest thing out there in terms of potential, start ups, etc. In fact I'd rather switch from quant to that than the other way around...

[chiral3]

Picked some comp bio types doing protein folding (heavy on knots, combinitorics) and some neurology.

[katastrofa]

"I wonder why you want to switch? Computation biology/biotech is probably the hottest thing out there in terms of potential, start ups, etc. In fact I'd rather switch from quant to that than the other way around..."



Prob. because, even though it's hard to get an entry-level quant job, it's even harder to get a postdoc job in a good uni. My SO is going through the process and they told her they got over 300 applications for 1 position. Each of these with 3 reference letters from tenured professors. Go figure.

[LostBiologist]

Thanks for the replies guys,



Lys: I'm mostly doing simulation of smaller networks, as opposed to the larger meta-analysis kind of work you're thinking of, so SDEs are pretty essential. Depressingly, while the current work is really interesting, and I'd be reasonably confident of getting a post-doc somewhere, it doesn't look like a particularly attractive career option for the long term (low pay, constant need to apply for more grants, decreasing chances of ever actually running your own lab), and the number of labs doing the exciting, interesting stuff is pretty damn small.



katastrofa: Yeah, it's rubbish

[Tradenator]

pharmaceutical company?  IBM?

[Jim]

@LostBiologist - If you've done this sort of thing, survival analysis math is being used more and more for modelling credit and prepayment events.  A biologist coming to the financial world would have a lot more credibility if he/she brought experience with heterogenity and multi-state models than with option pricing math.

[LostBiologist]

@Jim, That's interesting. I haven't done much survival analysis by that name, but the stuff I'm doing at the moment works on the same principals: modelling the time till particular biochemical interactions (using stochastic petri nets, if you are interested), and so on: if anything, I would have thought survival analysis was a bit basic to be an advantage in looking for a position. I guess Risk would be the key area that would be interested in this kind of thing, right? Are there other areas that use it?



Thanks for your input

[Jim]

@LostBiologist - Hazard rate (and other time-until-event) models are still quite common in MBS and asset-backed security pricing.  The more recent work involves time-varying heterogenity (i.e. modelling clustering or contagion affects), recurrent multi-state models (i.e., curing with relapses),  and using banks of computers (i.e. computing "clouds") for calibrating models on large amounts of data.

[chiral3]

No offense to anyone in particular, but these kinds of conversations tend to drive me nuts. (Junior) resumes I always immediately disqualify are those that say 1) built monte carlo model, 2) built regression model, 3) authored paper on hidden markov model.... It is analogous to saying 1) swung hammer, 2) used screwdriver, ..., instead of 1) built 5 room house utilizing hammer and screwdriver for construction. No broader context or understanding, just hammers and nails. Would you want to go to rates? equity? risk? some dead-language like convertibles or cdos? The tools are incidental not primary. If you have a bio PhD you are going to be hired as a lifer crank-turner who will leave in a few years or you will be able to leverage your ability to work on difficult problems with little guidance and influence your surroundings and have a potentially great career. Mark's book is great but it shouldn't be your only guiding light.



BTW - the job market is competitive and tight at the junior level and almost non-existent at the senior level compared to several years ago, likely to not return anytime soon. The quality people are staying put / being retained and the less-than-quality people are free and exponentially increasing their linkedin connections. Think long and hard about why you really want to move to this space. A different frying pan is still a frying pan.