6 ms·

Financial Software Projects (C++) - NYU Fall 2011

90 points by zura 15y ago

alanpca 15y ago
Oh man, I wish this was available when I was in school. This field has always interested me, but it seems that the bar of entry is super high (trading account).
- pnathan 15y ago
  There are a variety of companies that offer some level of programmatic access for trading.
  - Nrsolis 15y ago
    An Interactive Brokers trading account with a minimum 10K USD deposit will get you access to their trading API. If you can deposit 25K, you'll be OK with pattern day-trader status and gain 4x leverage for your account. They also offer a paper-trading account for testing purposes. They give you $1M in funny money to trade with to test your algos.
    - darkmethod 15y ago
      Can you please provide more information on this? A link would be fantastic. Thanks.
      - Nrsolis 15y ago
        http://www.interactivebrokers.com/ibg/main.php http://www.interactivebrokers.com/ibg/main.php http://individuals.interactivebrokers.com/en/pagemap/pagemap_APISolutions.php http://individuals.interactivebrokers.com/en/pagemap/pagemap...
    - steve8918 15y ago
      Even cheaper, if you want to trade futures, the margin requirements are even lower, $500-1000 minimum, and you can use Ninjatrader as your trading platform for certain brokerages. Ninjatrader has a pretty good C# interface to trade with. You have to pay for Ninjatrader if you want to trade, but you can sim trade and write your own trading strategies and backtest for free until you're ready to trade for real.
dustineichler 15y ago
I'm kinda finance geek, but I know nothing about this field programmatically. This looks amazing. Anyone have any other links worth looking at?
- Flemlord 15y ago
  QuantLib is a great open source finance library. http://quantlib.org/index.shtml http://quantlib.org/index.shtml
- rluhar 15y ago
  You may want to check out the forums at wilmott.com or the forums at nuclearphynance.com A lot of work in finance is also done in statistical programming languages like R. Some firms have also made big investment in vector processing / APL type languages like q (or k) and time series databases like kdb+. Event processing / correlation platforms and programming is also gaining a lot of traction in finance. Progress Software's Apama platform, Streambase, and a bunch of other start ups are competing in this space as well. Technology in finance is huge. It goes from building pricing models for derivatives and bonds (post under discussion), high frequency trading (FPGAs, event correlation,etc), to trade processing (high throughput transaction processing), to web app development (retail trading platforms, ebanking).... I could keep going on and on. It is wide field and if you are interested in working in technology in finance, it is safe to say you could find some nice in which you could use your skills. Obviously pay grades, job quality, etc. vary..
  - dustineichler 15y ago
    I'm only familiar with building pricing models. Thanks for the quick list and links. I kinda want to get into this. Sans hat tip.
AznHisoka 15y ago
I took a similar class a few years ago at NYU. All I can say is classes always look better at paper than they do in real-life. I learned absolutely nothing of substance, and found it a waste of time. But YMMV.
JumpCrisscross 15y ago
As someone who was a trader on a derivatives prop desk I would suggest those interested in finance with solid coding skills skip being a 20-year old's peon and: (1) Join a quant fund, e.g. Renaissance, Two Sigma, Citadel, etc., or, (2) Join a finance tech start-up, e.g. Palantir, Addepar, SecondMarket, Wealthfront, etc. who are liable to turn the industry over in the near future Both options require you be more than a "developer" (in the Wall St sense). Coding is a means by which to express your edge in data analysis and algorithmics. A coder who knows some finance is a "dev". A dev who knows statistical methods is a "quant". A quant who can engineer new trades is a Master of the Universe (and one to put the phone monkeys to shame at that). Note about the course books: this explains a lot about our devs. They have you implement VaR and other theoretically sophisticated programmes while giving a ram-jet flyover of finance that would leave you helpless with more the intricate articles in The Economist; it's the equivalent of learning CS theory with no coding experience or math through memorisation. Note about note: I'm not criticising the course per se. It does a fine job of preparing a CS student for being a developer on Wall St. I'm targeting this more to the HN community, who I feel would be wasted talent in that role.
- larsberg 15y ago
  Make sure to investigate the firm and, in particular, their enforcement of non-competes before you join to ensure you are comfortable with their terms. One of the three in your bullet (1) has a long history of enforcing (and backing up with Full Legal Might) an 18-month non-compete on even somewhat competent developers and traders.
  - JumpCrisscross 15y ago
    Or an effective lifetime non-compete if you're talking about Renaissance. Then again, their flagship fund, which is now only open to employees, regularly beats 50% annualised returns after fees.
- dustineichler 15y ago
  What other options are there than being a quant?
chollida1 15y ago
I'm not sure what a developer can learn about software engineering from Soul of a New Machine by Tracy Kidder but it is an awesome book. I'd recommend people read it for pleasure.
jordhy 15y ago
This is actually one of the hottest fields in C/C++ software development. Either high frequency trading or machine learning based transactions are in very high demand. I wish I could take this class.
Drbble 15y ago
For everyone lamenting not being able to take this class. Taking the class is the least important aspect. Going to college or equivalent and learning and practicing how to think hard are all you need classes for. Anyone who has completed a degree in any hard science /math /engineering can learn this material from the books and websites and forums.

90 points by zura 15y ago

Oh man, I wish this was available when I was in school. This field has always interested me, but it seems that the bar of entry is super high (trading account).

pnathan 15y ago
There are a variety of companies that offer some level of programmatic access for trading.
- Nrsolis 15y ago
  An Interactive Brokers trading account with a minimum 10K USD deposit will get you access to their trading API. If you can deposit 25K, you'll be OK with pattern day-trader status and gain 4x leverage for your account. They also offer a paper-trading account for testing purposes. They give you $1M in funny money to trade with to test your algos.
  - darkmethod 15y ago
    Can you please provide more information on this? A link would be fantastic. Thanks.
    - Nrsolis 15y ago
      http://www.interactivebrokers.com/ibg/main.php http://www.interactivebrokers.com/ibg/main.php http://individuals.interactivebrokers.com/en/pagemap/pagemap_APISolutions.php http://individuals.interactivebrokers.com/en/pagemap/pagemap...
  - steve8918 15y ago
    Even cheaper, if you want to trade futures, the margin requirements are even lower, $500-1000 minimum, and you can use Ninjatrader as your trading platform for certain brokerages. Ninjatrader has a pretty good C# interface to trade with. You have to pay for Ninjatrader if you want to trade, but you can sim trade and write your own trading strategies and backtest for free until you're ready to trade for real.

dustineichler 15y ago

I'm kinda finance geek, but I know nothing about this field programmatically. This looks amazing. Anyone have any other links worth looking at?

Flemlord 15y ago
QuantLib is a great open source finance library. http://quantlib.org/index.shtml http://quantlib.org/index.shtml
rluhar 15y ago
You may want to check out the forums at wilmott.com or the forums at nuclearphynance.com A lot of work in finance is also done in statistical programming languages like R. Some firms have also made big investment in vector processing / APL type languages like q (or k) and time series databases like kdb+. Event processing / correlation platforms and programming is also gaining a lot of traction in finance. Progress Software's Apama platform, Streambase, and a bunch of other start ups are competing in this space as well. Technology in finance is huge. It goes from building pricing models for derivatives and bonds (post under discussion), high frequency trading (FPGAs, event correlation,etc), to trade processing (high throughput transaction processing), to web app development (retail trading platforms, ebanking).... I could keep going on and on. It is wide field and if you are interested in working in technology in finance, it is safe to say you could find some nice in which you could use your skills. Obviously pay grades, job quality, etc. vary..
- dustineichler 15y ago
  I'm only familiar with building pricing models. Thanks for the quick list and links. I kinda want to get into this. Sans hat tip.

AznHisoka 15y ago

I took a similar class a few years ago at NYU. All I can say is classes always look better at paper than they do in real-life. I learned absolutely nothing of substance, and found it a waste of time. But YMMV.

JumpCrisscross 15y ago

As someone who was a trader on a derivatives prop desk I would suggest those interested in finance with solid coding skills skip being a 20-year old's peon and: (1) Join a quant fund, e.g. Renaissance, Two Sigma, Citadel, etc., or, (2) Join a finance tech start-up, e.g. Palantir, Addepar, SecondMarket, Wealthfront, etc. who are liable to turn the industry over in the near future Both options require you be more than a "developer" (in the Wall St sense). Coding is a means by which to express your edge in data analysis and algorithmics. A coder who knows some finance is a "dev". A dev who knows statistical methods is a "quant". A quant who can engineer new trades is a Master of the Universe (and one to put the phone monkeys to shame at that). Note about the course books: this explains a lot about our devs. They have you implement VaR and other theoretically sophisticated programmes while giving a ram-jet flyover of finance that would leave you helpless with more the intricate articles in The Economist; it's the equivalent of learning CS theory with no coding experience or math through memorisation. Note about note: I'm not criticising the course per se. It does a fine job of preparing a CS student for being a developer on Wall St. I'm targeting this more to the HN community, who I feel would be wasted talent in that role.

larsberg 15y ago
Make sure to investigate the firm and, in particular, their enforcement of non-competes before you join to ensure you are comfortable with their terms. One of the three in your bullet (1) has a long history of enforcing (and backing up with Full Legal Might) an 18-month non-compete on even somewhat competent developers and traders.
- JumpCrisscross 15y ago
  Or an effective lifetime non-compete if you're talking about Renaissance. Then again, their flagship fund, which is now only open to employees, regularly beats 50% annualised returns after fees.
dustineichler 15y ago
What other options are there than being a quant?

chollida1 15y ago

I'm not sure what a developer can learn about software engineering from Soul of a New Machine by Tracy Kidder but it is an awesome book. I'd recommend people read it for pleasure.

jordhy 15y ago

This is actually one of the hottest fields in C/C++ software development. Either high frequency trading or machine learning based transactions are in very high demand. I wish I could take this class.

Drbble 15y ago

For everyone lamenting not being able to take this class. Taking the class is the least important aspect. Going to college or equivalent and learning and practicing how to think hard are all you need classes for. Anyone who has completed a degree in any hard science /math /engineering can learn this material from the books and websites and forums.