022607

# 022607 - 01gxdx For some “complicated” function g Monte...

This preview shows pages 1–2. Sign up to view the full content.

Lecture 10: 02/26/2007 Recall: Segueing into discussing “Monte Carlo stuff.” Theoretical Tool : Strong Law of Large Numbers (SLLN). If x 1 , x 2 , x 3 ,… are iid random variables with finite (conversion) mean / expected value xbar, we have with probability 1, →∞ = = limn 1nk 1nxk x And variance of estimate -> 0 like 1/n. In fact, = ( = - ) varianceestimate E 1ni 1 nxi x 2 = ( ) variance xi n = ( )- E xi2 x2n Where we remember that the variance(xi) is the same for all i. Stanislaw Ulam, at Los Alamos, mid-40’s recognized (along w/ others) how new “computers” might give life to “statistical simulation”—rigorous basis=SLLN; computers would arbitrate actual calculations. Ulam coined name “Monte Carlo” for this collection of methods. First “interesting” problem: compute

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 01gxdx For some “complicated” function g. Monte Carlo approach: Make independent draws from a uniform distribution on [0,1] (lots of computer ways to do this.)—let the corresponding random variables ve x 1 ,x 2 , …, then by SLLN: = ( )→ →∞→ 1nk 1ng xi n 01gxdx =expected value of the rv g(x i ) wrt uniform distribution (Nb: {g(x i ): 1<=i<=0} is an iid sequence, (with y i =g(x i )), write common mean ybar= 01gxdx ) Computer’s roles: 1.) Generating random numbers 2.) Computing partial sums, n->infinity Another problem: (actually a special case of integration): compute area or volume of some complicated region D in some “easy” space V. <crash> <see onenote>...
View Full Document

{[ snackBarMessage ]}

### Page1 / 2

022607 - 01gxdx For some “complicated” function g Monte...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online