<t>I also use Java from time to time, but I dont see any speed improvements when using Javas built in generator over VBAs. Therefore I am tempted to just stay with Excel for Monte Carlo since outputting into a spreadsheet can sometimes be convinient. How would I go about programming something in Jav...

<t>I heard many on here bashing the vba rnd() uniform random number generator. By making a sample of 50000 I havent been able to find any real problems with it. It doesnt fail the tests for uniformity and it doesnt seem to show any significant serial correlation. It might not be 100% great but for m...

<t>Maybe it'll be helpful if I give a reference to the problem. It's problem 12.25 in Hull 5. ed. Or problem 13.27 in Hull 6. ed. The problem goes:Suppose that observations on a stock price (in dollars) at the end of each of 15 consecutive weeks are as follow:30.2 32.0 31.1 30.1 30.2 30.3 30.6 33.0 ...

Possibly a simple question but what is the formula for the standard error of a volatility estimate? I only know the formula for the standard error of the mean and I haven't been able to find any information on this anywhere.

I tried with a different random number simulator and now get much better results. I guess the quality of random numbers mean a lot.

<t>Just for the fun of it I have written a little vba program to simulate a GBM. My simulation has some unfortunate properties. When I increase the number of time steps in each year the mean drops to a level way below what it should be. I have looked at it again and again and I cannot find any probl...

Is it possible to get data from an Excel workbook into a C# array? Much data material comes on the form of an excel file so such an option would be very useful.

Hmm... Of course a bias can be problematic depending on the magnitude. In connection with this I have a hard time understanding what a bias "in the least significant bit" is referring to. Somebody enlighten me

Terror and Surutsu:It appears that the .NET generator isn't optimal. Are the problems merely of academic nature or will they cause problems in practise as well? The Mersenne Twister seems interesting and, as you said, quite easy to implement.

Okay, I see, thanks. But now I am getting curious - why would I implement my own uniform random number generator. Is there something wrong with the one found in .NET?

<t>QuoteWhere? In Random1.ccp see below. It should be mentioned that I only have limited knowledge of random number generation and is getting increasingly confussed. I just need to use a uniform distribution to get a normal one. double GetOneGaussianByBoxMuller(){ double result; double x; double y; ...

I assume that you're refering to the Box-Muller transformation. It looks interesting but I have some trouble understanding what is going on. Furthermore, it seems like I am getting a random pair of number instead of just one?

<t>Hmm... I have decided to code one myself. I figure the best option is to base it on the central limit theorem and rely on that the mean of the drawings from the uniform distribution will converge to the normal distribution. The question is what should my sample size be? Obviously I want it as sma...

<t>I am in the process of switching from Java to C#. Everything is going smoothly and I am generally not missing Java. However, I have one little problem. In Java the random number generator could create normally distributed random numbers. I am unable to find the same method in the .NET framework.H...

<t>What are your guys views on Java's performance and scalability? When I work with very large arrays (20000x1000) of data it sometimes grinds to a halt and wont even finish. Is this Java specific or what? Maybe the problem is that I try to draw all data to the screen. 10000x1000 is fine though. It ...