(This article by James Yardley is reproduce with permission from Australian Physics, vol.44 no.4, p140, November/December 2007).

Actually it started with dismantled toys, pyromania and the deaths of many, many insects, but this is possibly going too far back and is almost certainly not unusual among budding scientists in any case. More relevantly, having become interested in maths and physics (in particular) towards the end of high school, I almost got it right and enrolled in Electrical Engineering at Sydney University. This was in 1992 and fortunately by the end of my first year this concession to 'common sense' had been abandoned. I continued on to do honours (1995) and then a PhD (2001) in Theoretical Physics/Electromagnetism at Sydney with Prof. Ross McPhedran as my principal supervisor.

Although I greatly enjoyed physics and mathematics and had been convinced that I wanted an academic career, during my honours degree (4th year) it had begun to dawn on me that a career in theoretical physics was going to be quite a different experience to physics coursework. It was absolutely not going to be a case of daily insights and non-stop elegant results - at least for the majority (?) of researchers there was going to be considerable time spent chipping away at a fairly small patch of the coal face. It seemed that this was partly a result of the way success in academic work is measured and also due to the simple fact that a lot has already been done. Even though this narrowness would be mitigated by the healthy interaction of workers discussing their respective patches and of course teaching, it was still a blow to my hopes. Equally significant, was the likelihood that continuing in physics research would involve competing with very able people for very short contracts that didn't pay well in countries I might not want to live in.

A couple of photos: 1995 (physics) and 2007 (finance).

Early on in my university career I'd also developed an interest in the mathematics of gambling (The team room at one point becoming a den of trainee card counters) and from this the realization that there was much more to mathematical finance than book-keeping. Also, the stories of mathematicians and physicists (in particular) moving into Wall Street firms had begun to filter down - or at least I started noticing them. Most importantly of all, I had some friends from university who were already working as quantitative analysts and through this link I got some part time work at St George Bank doing mostly risk modelling.

This was at about the half way point in my PhD and meant that the most difficult part of selling out - landing that first job - was in my case quite straightforward. To be more specific, at this point I had only done half a course in the theory of option pricing and no formal statistics courses but with the typical range of physics, mathematics and computing completed by the half way point in a theoretical physics PhD, an enlightened employer might/did trust that the requisite finance could be learned on the job. This belief is justified by a long list of relevant skills a physicist might bring with them: a pragmatic approach to problem solving; the ability to reason about mathematical models in a non-mathematical way (Mathematicians don't necessarily understand diffusion just because they can solve the diffusion equation); the ability to understand and patch models when they fail; the ability to think heuristically; experience with real data; some programming skills; and hopefully the ability to communicate with both technical and non-technical people. This (albeit incomplete) list of relevant skills comprise reasons why physicists should make good quantitative analysts, but just possessing these skills does not obviate the need to actually learn some mathematical finance. Now that more courses are actually offered in mathematical finance it seems that employers are less likely to accept totally raw hard-science recruits with a view to teaching them the finance as it is needed. Worse still, even with a higher degree in physics and some mathematical finance courses under the belt there is a long-standing and partly legitimate fear among employers that such an academic person will be too aloof or otherwise unable to do anything relevant (i.e. profitable) to the business they are trying to run.

To be absolutely clear: there are no jobs in finance where former physicists or mathematicians sit in the lotus position while dirty faced underlings bring forth DE's to be solved. This is not to say that jobs in mathematical finance don't contain a significant academic component, but rather that an emphasis on quick and practical solutions is usually required

In 2002 I took a position in the Sydney office of an American firm (Susquehanna) where my role is primarily to use quantitative techniques to identify trading opportunities. As opposed to the quantitative methods involved in risk management (Although I can't attribute it, I recently heard mathematical finance (as applied to risk management) described as being like a safety net under a tight-rope that's only there when you're not falling), the development of trading strategies often draws on a much wider range of mathematical techniques (eg time series analysis, signal processing, information theory etc) - not to mention numerous ad-hoc models intended to exploit a specific market inefficiency. Although still relatively scarce in Australia, the number of people in this type of role has grown on the back of the recent explosion in the number of hedge funds. Anyway, perhaps unexpectedly my exposure to the sadistic world of special functions during my PhD would prove helpful in this space. Firstly by providing a tool chest of functional forms with which to try and model any given effect, and secondly in helping me optimize these representations. Indeed, it is increasingly becoming the case in modern trading that being fast is almost as important as being smart.

Finally, to explain the qualification in the title, it is not uncommon for people who've come from various science backgrounds and have worked in mathematical finance to at least entertain the hope of one day returning to their scientific roots. Part of leaving physics in the first place meant accepting physics as an interest, not a profession and as such a successful career in finance can be a very good thing in terms of allowing people to pursue other interests later on.

http://en.wikipedia.org/wiki/Econophysics - people who are determined to show that mathematical finance completely reduces to solved problems in physics.

"My Life as a Quant: Reflections on Physics and Finance", Emanuel Derman.