SERVING THE QUANTITATIVE FINANCE COMMUNITY

## Search found 81 matches

March 19th, 2003, 7:06 am
Forum: Programming and Software Forum
Topic: C & C++
Replies: 57
Views: 193302

### C & C++

Please give an example of hour long evaluation.
March 19th, 2003, 6:40 am
Forum: Programming and Software Forum
Topic: Matrices in C++ / C#
Replies: 14
Views: 190685

### Matrices in C++ / C#

<t>That depends on what you want to do with the matrices.If you only want to do basic stuff like addition, multiplication,...but no eigenvalues, eigenvectors,...you do not have the KAI or Intel compiler oryou do not care about the fastest possible execution --- do it yourself.It's often faster to wr...
March 19th, 2003, 6:06 am
Forum: The Quantitative Finance FAQs Project
Topic: What is Girsanov's theorem, and why is it important in finance?
Replies: 79
Views: 215253

### What is Girsanov's theorem, and why is it important in finance?

What happened to the aliens?
March 19th, 2003, 4:17 am
Forum: The Quantitative Finance FAQs Project
Topic: What is change of numeraire and why is it useful in finance?
Replies: 13
Views: 242637

### What is change of numeraire and why is it useful in finance?

<t>On a fundamental level numeraires are necessary to reduce general price processes to local martingales.Cash prices will not in general be a local martingale in any probablity because of inflationary drift upward(the money illusion).Cash prices divided by the numeraire can become local martingales...
March 19th, 2003, 3:52 am
Forum: The Quantitative Finance FAQs Project
Topic: What are low discrepancy numbers?
Replies: 8
Views: 195881

### What are low discrepancy numbers?

<t>Every Monte Carlo simulation computes an expectation E(X) as an averageE(f(X)) = (f(X1)+f(X2)+...+f(Xn))/n,where the Xj are independent observation from the distribution of X. This is justifid by the Strong Law of Large Numbers (SLLN)E(f(X)) = lim_{n->oo} (f(X1)+f(X2)+...+f(Xn))/n (*)Usually this...
December 2nd, 2002, 10:23 am
Forum: Technical Forum
Topic: BGM implementation
Replies: 12
Views: 197212

### BGM implementation

mj,I was in error about my drift term. It is quadratic not linear in n -- regardless of the number of factors --it is an unsophisticated implementation.This explains the limited gains which I see when reducing the number of factors.
November 12th, 2002, 7:22 pm
Forum: Programming and Software Forum
Topic: Pointers or References?
Replies: 13
Views: 190137

### Pointers or References?

Why not writeA=B; A*=C; A+=D;where the operators *=, += write into "this" and return "this" and also delete any workspace if necessary.
November 8th, 2002, 9:50 pm
Forum: Technical Forum
Topic: BGM implementation
Replies: 12
Views: 197212

### BGM implementation

<r>mj,It was number of factors times number of rates.I'll check this again to see why I am not getting this nice increase in speed.To drop from 50 seconds to 16 seconds is significant.I have a question: the Libor Market Model seems to be suitable for risk management of treasuries.Is there already so...
November 8th, 2002, 4:03 pm
Forum: Technical Forum
Topic: BGM implementation
Replies: 12
Views: 197212

### BGM implementation

<t>I concur with pj.Reducing the number of factors makes the implementation slightly more complicated since you have to diagonalize the covariation matrixand kill eigenvalues. I tried this in hopes of speedup and found out that virtually no speedup was observed unless the number of factors was reduc...
November 4th, 2002, 11:11 am
Forum: Programming and Software Forum
Topic: C++ v. Matlab v. Mathematica
Replies: 53
Views: 194502

### C++ v. Matlab v. Mathematica

<t>DoofusMaximus,I am not a professional quant developper but I would love to do this and will be looking for a job soon.So far as I can tell Java is not used for quant development. It seems to be C++ only.One argument in favour of C++ is that it is non proprietary. On the other hand development in ...
November 3rd, 2002, 3:24 pm
Forum: Programming and Software Forum
Topic: C++ v. Matlab v. Mathematica
Replies: 53
Views: 194502

### C++ v. Matlab v. Mathematica

<r>If you like Java please check out <URL url="http://martingale.berlios.de/Martingale.html">http://martingale.berlios.de/Martingale.html</URL>(a Java library for Monte Carlo simulation).Before I moved from C++ to Java I checked the speed of Cholesky factorizations (identical code)in both Java and C...
November 1st, 2002, 9:46 pm
Forum: Technical Forum
Topic: Option Hedging
Replies: 26
Views: 192700

### Option Hedging

<t>There are some inaccuracies in the paper ("Weights for Discrete hedging"):1. page 4 from the top: states that every time new hedge weights are computed we can assume the current time is zero and the conditional expectationsare ordinary expectations. This is true only if the assset vector (includi...
October 28th, 2002, 3:36 pm
Forum: Technical Forum
Topic: Option Hedging
Replies: 26
Views: 192700

### Option Hedging

Paul,Thanks, that´s very kind.
October 27th, 2002, 9:57 pm
Forum: Technical Forum
Topic: Option Hedging
Replies: 26
Views: 192700

### Option Hedging

OK, I´ll try to generate a PDF version on my Linux box as soon as I get back to the US.This should take until November 20th.
October 11th, 2002, 4:58 am
Forum: Technical Forum
Topic: Martingale and Hedging
Replies: 2
Views: 189475

### Martingale and Hedging

<t>The answer in the case of the call is yes and in the case of the general martingale quotient f/g (with g being the numeraire asset) is yes also provided s/g is a martingale also and we can represent the quotient f/g as a stochastic integralf(t)/g(t) = integral_0^t phi(u)d(s(u)/g(u)) (*)phi(t) is ...

GZIP: On