Lecture6 - Lecture 6 Mariana Olvera-Cravioto Columbia...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture 6 Mariana Olvera-Cravioto Columbia University molvera@ieor.columbia.edu February 6th, 2012 IEOR 4404, Simulation Lecture 6 1/17
Background image of page 1

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

View Full DocumentRight Arrow Icon
Generating random variates Assume that we are able to generate iid Uniform[0,1] random variates. Method 1: Inverse Transform Suppose we want to generate a r.v. X such that I X is continuous with CDF F ( x ) I F ( x ) is continuous and strictly increasing when 0 < F ( x ) < 1 (it can be just nondecreasing for F ( x ) = 0 or F ( x ) = 1 ) I F ( x ) has an inverse F - 1 ( x ) . Algorithm: 1. Generate U Uniform[0,1] 2. Return X = F - 1 ( U ) IEOR 4404, Simulation Lecture 6 2/17
Background image of page 2
Inverse Transform To see that X , generated by the inverse transform method, has the correct distribution note that: IEOR 4404, Simulation Lecture 6 3/17
Background image of page 3

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

View Full DocumentRight Arrow Icon
Inverse Transform To see that X , generated by the inverse transform method, has the correct distribution note that: P ( X x ) = P ( F - 1 ( U ) x ) = P ( U F ( x )) Since P ( U u ) = 0 , if u < 0 , u, if 0 u 1 , 1 , if u > 1 , then P ( X x ) = P ( U F ( x )) = F ( x ) IEOR 4404, Simulation Lecture 6 3/17
Background image of page 4
Example We want to generate an exponential r.v. with mean β . IEOR 4404, Simulation Lecture 6 4/17
Background image of page 5

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

View Full DocumentRight Arrow Icon
Example We want to generate an exponential r.v. with mean β . We start by writing the CDF of X : F ( x ) = ( 1 - e - x/β , if x 0 0 , otherwise. To invert F ( x ) : y = 1 - e - x/β e - x/β = 1 - y - x β = log(1 - y ) x = - β log(1 - y ) Therefore, F - 1 ( x ) = - β log(1 - x ) . So if U Uniform[0,1], then X = - β log(1 - U ) Exponential (1 ) IEOR 4404, Simulation Lecture 6 4/17
Background image of page 6
Note: If U is uniformly distributed in [0,1], then 1 - U is too, so we could have used X = - β log U IEOR 4404, Simulation Lecture 6 5/17
Background image of page 7

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

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 03/18/2012 for the course IEOR 4404 taught by Professor C during the Spring '10 term at Columbia.

Page1 / 23

Lecture6 - Lecture 6 Mariana Olvera-Cravioto Columbia...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online