bbot23
|
|
| Total Posts: 1 |
| Joined: Jul 2012 |
| |
|
Little bit of background: I graduated from Psychology about 2 years ago and subsequently worked in a prop shop for a year and a half. It was a great experience, mostly because I was successful for that short time period, but I quickly realized that the future of trading (prop or otherwise) was increasingly more focused on automation or semi-automation. I wanted to stay in the industry, but I couldn't shake off the feeling that my lack of programming and math/stats skills were holding me back.
Fast forward to today, I am two semesters into a B.Sc. in Computer Science. I am becoming incredibly more interested in statistics and machine learning so I'm tacking on a lot of discrete mathematics and probability/stats classes. Am I handicapping myself? Should I focus on math/stats as a major, with just 2-3 courses on OO programming?
Just to reiterate, I want to stay in the trading game. I have no real interest in Finance (I should mention that I attempted the CFA Level 1 and it didn't rock my boat), but anything that involves implementing clever ideas on a short time scale (HFT? Market-making?), applying behavioral principles, backtesting, forward testing, deploying automated strategies using rigorous, fundamental math/probability principles is really appealing to me. A company like Optiver has been on my radar, although I don't know many others like it. |
|
|
|
 |
sv507
|
|
| Total Posts: 133 |
| Joined: Aug 2010 |
| |
|
| i would say stats is probably the right major ( maths is prob too theoretical?). however it has to be said that you will still have a hard time finding a job.... there are plenty of other people with stats backgrounds. |
|
|
 |
|
What's good about the cs and stats together, is it's very useful and people with that background are quite trainable/worth training. If I were in your shoes, I would focus more on CS (or EE) as the major (if you can handle programming), and then take more applied stats like machine learning and statistical signal processing (usually in the EE department) to scratch your statistics itch.
In the early days of Paypal, that was the optimal type of engineer to hire and train, CS + applied stats and let them sink or swim. Now everyone wants these kinds of guys for all kinds of web analytics, so if you become disillusioned with finance (though it sounds like you have the trading bug), you also have other options. |
|
|
|
|
Aleph
|
|
| Total Posts: 54 |
| Joined: Jul 2006 |
| |
|
For people right out of university EE is probably the most widely considered 'safe' major for HFT. EE people are smart, have good technical and applied maths skills, and tend to be hard workers. Any further stats is a nice bonus, but I think it tends to be that the necessary stuff can be learned by those sorts of people on their own and their technical skills mean that they can iterate from rough and dirty solutions to something that works in reality at a very quick pace. Much preferred and more successful that those who spend a lot of time trying to find the 'perfect' solution.
With stats I think there's huge added value for knowing a bit, then it levels off quickly until you get into the really complicated stuff that's usually indicative of material past the undergrad (and even MSc) level. Even that stuff, in general, is perhaps not super large EV as while the payoff can be huge the probability of being able to apply it is much smaller than the basic stuff.
If I were giving advice to someone who wanted to maximize their chance of getting into a good prop shop and being successful I would recommend EE or CS with a few stat electives. Stay practical at all times! Write a lot of code and build things that work. The biggest problem with solely stats or finance focused people is a lack of practicality. Engineers and applied-CS tend not to have that problem by virtue of their work with real stuff that breaks and has bugs. |
|
|
|
