Any QuantLib experts here?

[earth_imperator]

You don't need QuantLib for that. In fact, you can easily simulate stock prices with an underlying GBM engine within excel.

Exactly, he could not even articulate why it must be in QuantLib. Probably some homework, lol. Not worth the effort.

The topic of this thread is QuantLib. Then why do you suggest I should use something else? Illogical you are, you pseudo expert

 

[M.W.]

Because tools are just tool. Are you here to solve a problem or you wanna be religious about tools?

The topic of this thread is QuantLib. Then why do you suggest I should use something else? Illogical you are, you pseudo expert

 

[earth_imperator]

Because tools are just tool. Are you here to solve a problem or you wanna be religious about tools?

I'm doing an exact math research, not looking for a questionable wishi washi quicky "solution" of yours.
You seem not have grasped the problem in the first place: the problem is to test whether QuantLib's GBM implementation is correct or not. Then how will it be useful to use something else but QuantLib? You are illogical, madam!

 

[M.W.]

What are you even blabbering about. You are as scientifically rigorous as a screw. GBM is a well defined process, driven by BM, to simulate all sorts of random events and their evolution. The math is in the underlying process, not in the tools that implement it. Do whatever you need to do, choose whatever tool you want. Your comment, however, is stupid. I hope you realize that as you progress and learn.

I'm doing an exact math research, not looking for a questionable wishi washi quicky "solution" of yours.

 

[earth_imperator]

Just an update:
Using the above said QuantLib GBM code I get very very weired results.
IMO the GBM in QuantLib can never ever be correct.
WIll continue my tests tomorrow.
Will also post the code here for verification & reproduction of the results by others.

Ok, just posted the C++ test code incl. my findings (results):
https://www.elitetrader.com/et/threads/simulating-stock-prices-using-gbm.375533/page-4#post-5861368

 

[ph1l]

@ph1l, thx, can you or your BingChat perform the said test here?
...
Ie. the challenge is: Does your GBM algorithm pass the "68–95–99.7 rule test"?
They are 3 tests, but the first (ie. the 68.27% test) just suffices to pass.

That seems to be above its pay grade.

In the context of the Geometric Brownian Motion (GBM) model created using QuantLib with the given parameters of S0=100, t=1 year, volatility s=0.3 (ie. 30%), and drift u=0%, it is important to note that GBM is a continuous-time stochastic process where the logarithm of the randomly varying quantity follows a Brownian motion with drift. The distribution of the logarithm of the asset price at any given time follows a normal distribution, so it is possible to apply the empirical rule to analyze the probability of the asset price falling within certain intervals.

However, without running simulations or performing further analysis on your specific GBM model, it is not possible for me to determine whether or not it passes the "68–95–99.7 rule test". I hope this helps clarify things! Let me know if you have any further questions.

 

[earth_imperator]

That seems to be above its pay grade.

Bing Chat said:
In the context of the Geometric Brownian Motion (GBM) model created using QuantLib with the given parameters of S0=100, t=1 year, volatility s=0.3 (ie. 30%), and drift u=0%, it is important to note that GBM is a continuous-time stochastic process where the logarithm of the randomly varying quantity follows a Brownian motion with drift. The distribution of the logarithm of the asset price at any given time follows a normal distribution, so it is possible to apply the empirical rule to analyze the probability of the asset price falling within certain intervals.

However, without running simulations or performing further analysis on your specific GBM model, it is not possible for me to determine whether or not it passes the "68–95–99.7 rule test". I hope this helps clarify things! Let me know if you have any further questions.

Nevertheless that's an impressive analysis by that AI bot. It understands the problem domain clearly and comes up with a suggestion better than most of the human experts...
And meanwhile the simulations have already been performed here, and the result confirms the made hypothesis that QuantLib's GBM fails to pass the above said empirical rule test (ie. the "68–95–99.7 rule test").

 

[traider]

beauty of quantlib is it can build an entire pricing engine for you meaning you bootstrap interest rate curves and it values every asset with it. But this comes at the cost of additional complication which makes it hard for beginners to pick up
But with power of gpt anything is possible if you dedicate to it

 

[blueraincap]

so you're looking for a gbm generator? what does this have to do with quantlib?

He forgot the formula and/or too lazy to write it