4 ms·

Saying "I want to learn Quant Finance" is like saying "I want to learn programming," in that it really all just depends. There's an impossible number of fields

by tdees40 13y ago

Saying "I want to learn Quant Finance" is like saying "I want to learn programming," in that it really all just depends. There's an impossible number of fields and sub-fields, and many people know one field intimately but may not know another at all (high-frequency trading and derivative pricing, for instance, have little overlap). Also, some of these jobs (i.e. high-frequency trading) are about 90% programming, whereas other jobs tend to be significantly more mathematical. If you can't narrow your interests beyond just the exceptionally broad category of quant finance, then learn enough to figure out what subcategory you want to jump into, and then let me know and I'll give you a reading list, if I know anything about it.

tdees40 13y ago
To get it started, subcategories include: 1) High-frequency trading. Very programming oriented as trade latency is everything. You'll need to be really good at making C/C++ code really fast. Math is somewhat less important than other fields. 2) Derivatives pricing. Usually working for banks/hedge funds, and use various methods (Monte Carlo, PDE, transform) to price options. Will need to know C/C++ and/or probably Matlab, C# (Excel is king here, so MSFT languages is a big deal), and Python is on the up and up. Will need to have a serious math background. This is more or less what I do. 3) Algo trading. Uses mathematical techniques to find patterns and execute trades. Probably works for a hedge fund. Programming languages vary by fund and include Matlab, Python, C/C++, Java (one uses Ocaml). Less theoretical, lots of playing around with data. 4) Quant risk. Usually works for a bank or hedge fund in the risk department. Responsible for things like building VaR models. Usually requires light programming (VBA, although serious languages help), and some lightweight math.
- kasey_junk 13y ago
  I'd add that if you are truly interested in HFT the time when you could make C/C++ software fast enough is either gone or very short lived. Much more valuable would be FPGA programming or other similar skills.
  - rgovind 13y ago
    For a layman, Can you please explain why FPGA skills are good to have? Some years, I could understand that processors were not fast enough so FPGA parallelism was useful...I think modern processors have caught up.
    - kasey_junk 13y ago
      FPGA cards are the easiest/cheapest way to get programs into the chip level. You can bypass OS level operations, memory/cache operations etc. That said, you probably could build bespoke machines out of modern processors that perform better (and someone might be doing this), but it would be very expensive. FPGA's offer a middle ground between that and software.
- jplewicke 13y ago
  Going to tack on: 5) Quantitative portfolio management. This usually involves managing a portfolio of stocks, either long-only or long-short. People usually come into this from either a heavy math/tech background or from doing academic finance research. Skills include portfolio optimization using linear/quadratic/etc. programming, machine learning, simulation and backtesting. Some companies that do this include: - Acadian Asset Management - AQR Capital - Goldman Sachs Asset Management - Janus InTech - MDT Advisers (discl: I work there) - Renaissance Technologies Institutional Equities Fund

by tdees40 13y ago

tdees40 13y ago

To get it started, subcategories include: 1) High-frequency trading. Very programming oriented as trade latency is everything. You'll need to be really good at making C/C++ code really fast. Math is somewhat less important than other fields. 2) Derivatives pricing. Usually working for banks/hedge funds, and use various methods (Monte Carlo, PDE, transform) to price options. Will need to know C/C++ and/or probably Matlab, C# (Excel is king here, so MSFT languages is a big deal), and Python is on the up and up. Will need to have a serious math background. This is more or less what I do. 3) Algo trading. Uses mathematical techniques to find patterns and execute trades. Probably works for a hedge fund. Programming languages vary by fund and include Matlab, Python, C/C++, Java (one uses Ocaml). Less theoretical, lots of playing around with data. 4) Quant risk. Usually works for a bank or hedge fund in the risk department. Responsible for things like building VaR models. Usually requires light programming (VBA, although serious languages help), and some lightweight math.

kasey_junk 13y ago
I'd add that if you are truly interested in HFT the time when you could make C/C++ software fast enough is either gone or very short lived. Much more valuable would be FPGA programming or other similar skills.
- rgovind 13y ago
  For a layman, Can you please explain why FPGA skills are good to have? Some years, I could understand that processors were not fast enough so FPGA parallelism was useful...I think modern processors have caught up.
  - kasey_junk 13y ago
    FPGA cards are the easiest/cheapest way to get programs into the chip level. You can bypass OS level operations, memory/cache operations etc. That said, you probably could build bespoke machines out of modern processors that perform better (and someone might be doing this), but it would be very expensive. FPGA's offer a middle ground between that and software.
jplewicke 13y ago
Going to tack on: 5) Quantitative portfolio management. This usually involves managing a portfolio of stocks, either long-only or long-short. People usually come into this from either a heavy math/tech background or from doing academic finance research. Skills include portfolio optimization using linear/quadratic/etc. programming, machine learning, simulation and backtesting. Some companies that do this include: - Acadian Asset Management - AQR Capital - Goldman Sachs Asset Management - Janus InTech - MDT Advisers (discl: I work there) - Renaissance Technologies Institutional Equities Fund