Physics, Maths & Philosophy: Wall Street Prospects

[Moss Pauly]

Hello I am currently thinking about uni options for next year. I am very geared towards physics, maths and philosophy. Through my many searches on the internet I stumbled upon the idea of getting a PhD and working on Wall Street. This idea grabbed my interest as one of my hobbies has been playing with arbitrage betting, I know this is not directly related at all but there is still an underlying passion for working with numbers about money.
What I was wanting to know is if anyone knows if they are and or would be employing people with this sort of background in 5-10 years. Also has anyone had any experience with this type of work here?

Thanks,
Moss

 

[twofish-quant]

Hello I am currently thinking about uni options for next year. I am very geared towards physics, maths and philosophy. Through my many searches on the internet I stumbled upon the idea of getting a PhD and working on Wall Street.

If you aren't particularly interested in physics, you'll find it a lot easier if you get a masters in either finance, business administration, law, computer science, or statistics (although you could go into those areas with an undergraduate physics degree.)

Wall Street is a wonderful place to work for physics geeks, but if you aren't a physics geek, then getting a Ph.D. just complicates things. Getting a physics Ph.D. is probably the worst way of getting into Wall Street.

This idea grabbed my interest as one of my hobbies has been playing with arbitrage betting

If that's your main interest, then getting a physics Ph.D. is seriously overkill. Also, the type of math that most physicists end up doing is more suited for derivative valuations, and people that do arbitrage on Wall Street tend to come out of econometrics, statistics, and EE departments. Now physics Ph.D.'s *can* learn new math quickly (which is why they get hired), but the type of math that people do in statistical arbitrage isn't what most physicists have deep knowledge of.

Now if your main interest is neutrino physics and you are looking for something interesting that keeps you fed, that's a different situation.

I know this is not directly related at all but there is still an underlying passion for working with numbers about money.

Screw passion. One thing that I like about Wall Street is that no one cares if you say that you are doing something for the money. The reason that people talk about passion so much in academia is that there is no money there.

What I was wanting to know is if anyone knows if they are and or would be employing people with this sort of background in 5-10 years.

No clue. One thing other thing that I like about markets is how unpredictable they are. One difference between an amateur and a professional is that a professional is much more likely to say "I just don't know." One thing that amateur traders tend to do is that they think they can outguess the market, whereas professionals deal with the fact that they have no clue which direction the market will go in.

This applies to labor markets too.

I can't tell you whether or not physics Ph.D.'s will be employable on Wall Street in five to ten years. I might be able to tell you something about what things are going to be like in one to two years. But that's what makes the game interesting.

Curiously one thing that makes long term future prediction impossible is that the predictions will change the future. If everyone was convinced that having a physics Ph.D. on Wall Street would make you a killing, then everyone does that then everyone gets physics Ph.D.'s and you have a massive oversupply (which is basically what happened with real estate). Also this happens on smaller scale. If you sell $100 of Exxon stock, you aren't going to change the market.

If you try to sell $10 million of Exxon stock, you are going to have to think very hard about how to do it, because you will flood the market and lose your shirt if you do it wrong, and a lot of what arbitrage trading involves are people that are trying very hard not to get arbitraged. There's more money playing "anti-arbitrage" than there is playing "arbitrage."

The other thing that is interesting is the idea of price. You *think* you know what the price of Exxon stock is, but you don't. You know what the price was 5 seconds or even 200 milliseconds ago, but that's different from the price "now" (if there is such a thing). Something that amateurs need to realize is that if someone quotes you a price, and you buy or sell using that price, you just got screwed.

Also has anyone had any experience with this type of work here?

For me, money is nice, but I can't guarantee you or myself money. What I can guarantee is adventure. If the world financial system blows up again, I'd rather be somewhere that I can see it happen first hand, rather than read about it in the newspapers.

Ummm... Yes...

 

[Mute]

Curiously one thing that makes long term future prediction impossible is that the predictions will change the future. If everyone was convinced that having a physics Ph.D. on Wall Street would make you a killing, then everyone does that then everyone gets physics Ph.D.'s and you have a massive oversupply (which is basically what happened with real estate). Also this happens on smaller scale. If you sell $100 of Exxon stock, you aren't going to change the market.

One of my favourite anecdotes about applications of statistical physics is that the Black-Scholes equation only fits the data back to a certain point in time. Going back further than this point, the data and the model don't really match up very well at all - the reason being that the point in time when the data and model agree is after the model was derived! I'm not sure how true the details of the anecdote are, but I find the idea of one's model interacting with the system it is supposed to be modeling to be positively delightful.

(Sorry for slightly-off-topicness)

 

[twofish-quant]

One of my favourite anecdotes about applications of statistical physics is that the Black-Scholes equation only fits the data back to a certain point in time.

The standard Black-Scholes equation doesn't fit the data at all after 1987. What happened was that Black-Scholes assumes that you can sell stock instantly at a given price. And before 1987, people assumed that was true. In 1987, it became obvious that that wasn't true, and people started pricing in "gap risk" and that changes the numbers.

I'm not sure how true the details of the anecdote are, but I find the idea of one's model interacting with the system it is supposed to be modeling to be positively delightful.

And it's something that you have to worry about.

The way I explain how models are used in Wall Street is with bananas. If I find that that I can buy a pound of bananas for 25 cents, how much should I expect to pay for two pounds of bananas. The answer is 50 cents, and most of physics based modelling on Wall Street is just what I did with differential equations and more math.

Now you might point out (correctly) that this is a vast oversimplification of reality, and you'd be right.

If I found out that I can buy a pound of bananas for 25 cents, but then I found out that I can buy a two pounds of bananas for 49 cents, that doesn't match the model. At that point I think about what is going on, and try to explain what is going on using the model as a reference, and try to figure out if there is money to be made. (For example, if I can *buy* two pounds of bananas for 49 cents, and *sell* one pound of bananas for 25 cents and then another point of bananas for 25 cents, I do it... If...)

The important thing is to *think* about what is going on, which is why physicists get hired for this sort of thing.

 

[Moss Pauly]

Thanks for the prompt detailed reply. I might have been a bit ambiguous in the first post, I really do enjoy doing physics and will probably do a science degree regardless of the prospects of working in a wall street situation. Arbitrage betting is by now means one of my strong interests in life just one of the many things I've been tinkering with lately. I just had a browse through the forums and read a fair few of the related topics to do with wall street with great interest. I guess I'll just see how the future unfolds :). I'm sure i'll have some more questions to ask you at a later date if that's ok? :)