# Math Career and networking tips for a physics/math major headed for finance?

1. Mar 30, 2012

### meanrev

Although it's my first choice, I don't see high hopes of having a career in academia. As such, I've decided not to put all of my eggs in the same basket, and instead first set myself up for a backup option in the industry.

The issue is that my interest lies with providing smart grid infrastructure for trading exchange and brokerage servers, designing retail- and institutional-end quantitative risk management software (nothing fanciful... think about something that can keep up with risk measures, hypothesis testing of modeled loss distributions on a rolling basis, and maybe with F# API - because I think it's fast and neat) and market-neutral and market-making trading algorithms. What I know is that the barriers to entry to these areas are very high, and I don't know where to start.

Now, I come from a poor family and I'm about to finish my second year at a small liberal arts college. I've done some of my part, by knowing nearly all of my schoolmates on a first-name basis, and getting elected as the president of the physics society recently. But it seems like none or few of my schoolmates have immediate connections with any high net wealth individual. (Well, at most someone's-dad's-friend is rich, that's about it.)

Option 1: Start on my own
I have about $10,000 in risk capital, but few friends I'd rely on for a startup. I don't even think I'm anywhere near good enough. [Not sure if I got it right: Yes, I've worked it out and realized it's enough for speculatory trading in futures (well, for paying intraday margin on 1 contract with enough to absorb maximum drawdown/consecutive losses). But I'm not keen on this because I view trading as a kind of gambling where there is asymmetric information and the rules of the game are not known - and gambles never seem to be in my favor. *shrugs* Which is why I'd only take on market-neutral positions or at least some kind of hedged pair. I have written and backtested/simulated some basic trading algorithms with unfiltered tick data... well, they sort of pick up on the volatility premium and use option spreads on the underlying instead of naked stocks where appropriate for a classical mean reversion strategy, and I mix it up with some momentum strategies, producing 5-6 Sharpe ratio together - I think the slippages are just going to make this unfeasible - though I'm confident of getting it to 2 digit. But then again, I don't even have enough for such an account trading in equities/ETFs, and this type of returns can't feed me because I'm not managing a huge fund/insurance company.] Option 2: Conventional paths It seems like a combination of my school-branding, declared majors (and admissibly, I probably know much less about finance than the kids who view this as their 1st choice career) is getting in the way of the intern -> analyst-> associate... path into a bulge bracket investment bank/proprietary trading firm that would be the most common entry into this field. (Moreover, I won't get any direct work of the infrastructure/software/market-making algorithms kind that interests me - but this is a non-factor as I'd accept such an internship offer any time.) Alternatively, I've considered focusing on academia, then getting out with a graduate degree that could get me hired in a market-making, high frequency trading firm. But I can't afford grad school - my mother is getting old, and she will run out of life savings from keeping me in college by the time I graduate. And I can't justify working 10-12 hours a day on paltry pay when I am fit to work and pay for her living expenses. I want to help her out as soon as possible: What happens if I go down with a disabling condition, lose my life in a traffic accident etc. during those long years of grad school? Question Is anyone familiar with my depressing situation and the financial industry? Haha. So, how should I network my way into the industry this summer? 2. Mar 30, 2012 ### chiro Hey meanrev and welcome to the forums. I just want to say a few thoughts that were on my mind. With regards to developing software, I just have to ask: how much software development (doesn't matter if you've written a few procedures you have put in a library or whether you have written an entire application) have you done? If you haven't done all that much, I would suggest you be aware of what it's really like to develop even a small library and the pain that comes with development especially when you are developing complex things that require you to use lots of different platforms integrated together. I won't talk about all the other issues because I don't see the point in going through those if you haven't had any solid exposure to developing software of any scale. I think you have already stated that you know the speculative nature of what you are doing which I think is good because having that awareness if you use it right could make you a lot less reckless than it otherwise would. Also I'd be interested in how much of your initial investment of 10 grand is tied up in fees or transaction costs of any kind because it might be the case that it's just not worth putting that kind of money if the fees end up taking a huge chunk of your investment (to me it's like throwing money for the 'privilege' of being allowed to gamble which again to me is stupid). The only thing I have to ask is if you are trading, then what do you know about the market that you are trading in? I'm not talking about mathematical modelling, but more so on the history of the market and how to assess the different risks that have been highlighted by events in the history of the market. For example in your market, what is the liquidity risk like? What are the major events for that market that have occured over the last half a century to a century and what are the lessons learned in terms of risk assessment and its implications? How is the market distributed? (In other words what kind of economic model is involved? Is it a highly distributed market or is it a highly non-distributed market?) In response to this, could someone with enough money or a strong position move the market tomorrow? Where is the bulk of the trading done for your particular market done and more importantly by who? (Largely computer trading through algorithmic procedures? Most trades done by large banks?). Also if you had to understand valuation (I mean real valuation that associated with proper price discovery and not something that is written on a ticker tape or an index), then how would you really value that thing? Does the market history support your statement? The point of the questions above is to realize the kinds of uncertainties that are not and can not be quantified by a distribution effectively. If you are not considering the kinds of questions above, then in my mind you are doing more gambling than actual trading. Mathematics is of course very useful and it's good to see simulations and models that are calculated using algorithms that have well understood assumptions, but unfortunately mathematics doesn't really explain IMO effectively, the understanding of the process. You could have some of the greatest pattern recognition algorithms and models that see things that a thousand of the best analysts would never see, but the computer doesn't really know how to actually interpret what they see. One final question I have for you: would you be willing to wait a long period of time after a lot of studying, analyzing, and waiting for an opportunity to present itself so that you could pounce on an opportunity and make a decision that has been in careful planning with a lot of diligence as a result of a lot of effort over say a trade that has been made without that kind of preparation, analysis, or effort? 3. Apr 3, 2012 ### twofish-quant There are lots of ways of getting in. It depends on what you are trying to do. If you are trying to start your own company, then it's going to be incredibly difficult. I along with everyone else will tell you that's it's impossible, you won't listen to us, and instead if you might pull off a Bill Gates or Michael Dell. But if you have that level of entrepreneural ambition, then you'll know better than listen to anything I say. Now assuming you aren't Bill Gates. It's not hard to get a job working on financial software. Since this is a "geek board", one way of doing it is to just be very good at statistics or computer programming, and then get a job at a bank or hedge fund. If you look at a typical bank, you have *hundreds* of people babysitting the trading systems, which means lots of jobs. The other problem is that those high net worth invididuals are likely to want to get their money managed by other people. One other thing to point out is that there are tons of financial jobs that are serving ordinary people. Everytime you put an ATM card into a machine or log into your online brokerage, there are dozens of programmers responsible for that. You seem to have it together. What kills most trading strategies is transaction cost. Professionals make money by being the transaction cost. With$10K, you are probably going to have a higher return just putting it into an index fund and forgeting about it.

Perhaps, but that's a different group. For the people that are reading this group, the most likely ways in are through computer programming, physics, mathematics, and/or statistics. One thing that I've found to be very useful in getting a job in finance is that I don't want a job in finance. I want a job in astrophysics. Now if I wanted a job in finance I'd be competing against everyone else that wants a job in finance. I want a job in astrophysics. I randomly wandered into finance because it was the most interesting thing that I could find, but because I don't care that much about finance, they have to give me more money to keep me to stay.

Physics and math graduate students make money by being indentured academic serfs. It's really a good deal. The other thing is that if you want to go that route, then it's not a good idea to have a well defined idea of what things will be like. It's possible that in a decade high frequency trading could be dead.

If you want to go through the standard finance student route, you are in the wrong forum. However, if you want to do something physics/math related, you are better off getting an REU in something non-financial.

4. May 5, 2012

### meanrev

@chiro: Thanks for your critical assessment. I have thought of some of the questions you've posed. I have answers to a few (the market participants/liquidity risk), but the others did leave me thinking. [Well currently I am no different from a speculative trader in DAX futures, there's definitely institutional and black box manipulation of the prices, since I see very quick processes driving the cointegration of its prices with the S&P 500 / EuroStoxx - various kinds of arbitrage/hedge going on, and the volumes on Fridays suggest that there's institutional weekly rebalancing. I also see 500+ lots in a single order getting chewed up instantaneously, which has to be driven by some large players with algorithms (500 lots would translate to something like 85 million euros worth of derivatives). There's not much liquidity risk for my portfolio sizes, but at I expect slippages of 1-2 tick sizes at certain hours when the bid-ask spread is wide.]

After reading through your post, I agree with you that my biggest weakness is that I don't consider myself to have enough preparation. I thought about it carefully, and set myself the goal of preparing myself aggressively for this course.

@twofish-quant: I totally agree. These are challenges that I can't mitigate at this stage. I don't have to capital to engage a prime brokerage which would make the transactional costs viable, so I decided to change my business strategy. On your urging (and of those around me), I applied to a few programming/quantitative analyst positions out there (Bloomberg, Knight Capital, DE Shaw etc.) but I was too late - I consider it a huge mistake. But the mistake did push me to come to a huge decision, one I should have made earlier.

I dropped out of college to pursue my goal. (I'll write a post later.)