• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

what do you think about Vb.net

alain said:
the whole NASDAQ infrastructure is written in Java so there is some sort of a contradiction here. NASDAQ is not the only one, the LMAX exchange is also completely written in Java. The whole speed bottleneck is something very rare nowadays (it was 3 years ago). Check the LMAX architecture and the Disruptor pattern (http://martinfowler.com/articles/lmax.html) when you get a chance. Also, read this blog http://mechanical-sympathy.blogspot.com/ . It's written by one of the authors of LMAX. This is high performance java at its finest.

I haven't done heavy duty java for at least 2 years so I'm sort of disconnected from the cutting edge but the problems that you are having, have been solved.

As an aside, the JVM is growing. It's extremely stable, fast and proven. Java has sort of stalled but new languages on the JVM are evolving like Scala and Clojure.
Click to expand...

indeed Java does the job well here but the complexity lies at the matching engine. The exchange simply accepts data from a number of designated points like brokers and clearing firms and store them in a database. The matching engine just needs to match the current trading prices for a particular ticker and close the deal between two counter-parties, along with some rules. Which is why exchanges have super huge data centers the size of Aurora - to allocate more hardware horsepower to lesser data points. And why exchanges don't dabble in analytics.

Java is of course used at the enterprise level but working in the high-frequency domain i have problems telling my bosses to choose java over c++. A few microseconds delay is unacceptable. Our goal is to go low in the kernel level while beating the other HFT software out in the market. Sometimes the exchanges experience a few seconds of round-trip delays in super volatile moments and that could mean a win or lose for some traders' livelihood.
 
Phume said:
Not sure yet. I just like to have option.
Quant, Analytical or build financial model
What other language do you suggest for Finance? JAVA?
How about for data science? Python,R?
Click to expand...
Like R, Python does have SciPy and NumPy packages which you can easily integrate for scientific computing and array manipulation.

Matlab does have lots of financial functions built in like black scholes formula and trees for option prices.
 
Back
Top