Does Matlab have a library for Monte Carlo integration

2 ビュー (過去 30 日間)
Radoslav Vuchkov 2021 年 6 月 29 日
コメント済み: Radoslav Vuchkov 2021 年 6 月 29 日
Hello,
I have been doing some reading and I cannot seem to find a simple to use matlab library to do Monte Carlo integration in n dimentions. For my needs I want to integrate a function over a n dimentional square.
Thank you very much for the inputs and ideas.

サインインしてコメントする。

回答 (1 件)

John D'Errico 2021 年 6 月 29 日

Do they have a "library"? Not really what I would call a "library". Maybe you might. But why would they? They offer random number generation functions, which is all you need. In this case, all you need is the function RAND. RAND can generate sets of points that lie uniformly inside an n-dimensional square.
1 件のコメント表示非表示 なし
Radoslav Vuchkov 2021 年 6 月 29 日
I understand what you are saying in fact I wrote the code below to do an integration over a square. However this converges really slowy and I was hoping for something a bit more safisticated if that makes sense. For example in python there is the qmcpy library where they use a bit more machinery for better/faster convergence.
sampleN = 10000000;
w = a + (b - a)*rand(d,sampleN);
MC = (1/sampleN)*sum(fun(w));

サインインしてコメントする。

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by