Thread: Python 3.0
View Single Post
Old 01-24-2009, 05:57 PM   #9 (permalink)
prescient63
Registered User
 
Join Date: Jan 2002
Posts: 1,921
-42 Internets
Quote:
Originally Posted by Etruscus View Post
I want to learn Python so I can write a program to calculate a specific type of financial asset that is priced multi-dimensionally. I need to be able to price using the monte-carlo method X2, but for the second run it would need a random walk generator (to simulate random walks in the interest rate).

The gui would need to allow for entry of up to 25 years, where an event has a certain likelihood of occurring every year, summing to 1000. This distribution looks more or less random, but it has a slight kurtosis. The distribution is empirical as well.

Excel just can't handle the level of computation that I need. I'm not on any kind of deadline, since they don't know I'm working on it. It occurred to me a couple weeks ago that calculating this asset by using the IRR iterative algorithm was leading to errors, since it always assumes anything reinvested would be reinvested at that specific IRR, and of course your discount rate changes as the market changes, so your output will always be erroneous.

For this project, learning 2.5 or 2.6 would be fine. But, I'm a computer gaming nerd at heart and I want to learn 3.0 to get a better understanding of what it means to actually run a large project like an MMO. I don't think there is a single computer game player in the market that hasn't said "I can do this better" and if they haven't, they are a console player Since 3.0 is the future, i think it makes sense to learn 3.0, even though very little is supporting right now.
You ever think about learning matlab (based in C from what I understand) instead. I know a lot of quants use it for prototyping and then code it in C++. I know some people in the office also like mathematica for this type of thing as well.

Last edited by prescient63; 01-24-2009 at 06:11 PM..
prescient63 is offline   Reply With Quote

 
Uberguilds Network