Latin Hypercube Sampling Example Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.

Slides:



Advertisements
Similar presentations
Space-Filling DOEs These designs use values of variables inside range instead of at boundaries Latin hypercubes, one popular space- filling DOE uses as.
Advertisements

Lectures prepared by: Elchanan Mossel Yelena Shvets Introduction to probability Stat 134 FAll 2005 Berkeley Follows Jim Pitman’s book: Probability Section.
Exponential and Poisson Chapter 5 Material. 2 Poisson Distribution [Discrete] Poisson distribution describes many random processes quite well and is mathematically.
Estimating a Population Variance
Random Number Generators Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
Histograms Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
MISC.. Simulating Normal Random Variables Simulation can provide a great deal of information about the behavior of a random variable.
Using random numbers Simulation: accounts for uncertainty: biology (large number of individuals), physics (large number of particles, quantum mechanics),
CS 351/ IT 351 Modelling and Simulation Technologies Random Variates Dr. Jim Holten.
Contemporary Engineering Economics, 4 th edition, © 2007 Risk Simulation Lecture No. 49 Chapter 12 Contemporary Engineering Economics Copyright, © 2006.
Simulating Normal Random Variables Simulation can provide a great deal of information about the behavior of a random variable.
- 1 - Summary of P-box Probability bound analysis (PBA) PBA can be implemented by nested Monte Carlo simulation. –Generate CDF for different instances.
Propagation of Uncertainty Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
Monte Carlo Simulation in Excel Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
Lecture II-2: Probability Review
Space-Filling DOEs Design of experiments (DOE) for noisy data tend to place points on the boundary of the domain. When the error in the surrogate is due.
Hydrologic Statistics
Copyright © 2010, 2007, 2004 Pearson Education, Inc. All Rights Reserved Section 8-6 Testing a Claim About a Standard Deviation or Variance.
Continuous Probability Distributions  Continuous Random Variable  A random variable whose space (set of possible values) is an entire interval of numbers.
1 Theoretical Physics Experimental Physics Equipment, Observation Gambling: Cards, Dice Fast PCs Random- number generators Monte- Carlo methods Experimental.
Copyright © 2010 Lumina Decision Systems, Inc. Monte Carlo Simulation Analytica User Group Modeling Uncertainty Series #3 13 May 2010 Lonnie Chrisman,
Sampling Methods  Sampling refers to how observations are “selected” from a probability distribution when the simulation is run. 1.
Contemporary Engineering Economics, 6 th edition Park Copyright © 2016 by Pearson Education, Inc. All Rights Reserved Risk Simulation Lecture No. 40 Chapter.
Use of moment generating functions 1.Using the moment generating functions of X, Y, Z, …determine the moment generating function of W = h(X, Y, Z, …).
Jake Blanchard University of Wisconsin Spring 2006.
A. Betâmio de Almeida Assessing Modelling Uncertainty A. Betâmio de Almeida Instituto Superior Técnico November 2004 Zaragoza, Spain 4th IMPACT Workshop.
McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies, Inc. All rights reserved. 1 Simulation.
SUPPLEMENT TO CHAPTER NINETEEN Irwin/McGraw-Hill © The McGraw-Hill Companies, Inc., 1999 SIMULATION 19S-1 Chapter 19 Supplement Simulation.
Sources of Uncertainty (from Morgan and Henrion) Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
IV. Inferential Statistics B. Confidence Intervals
ESD.70J Engineering Economy Module - Session 21 ESD.70J Engineering Economy Fall 2006 Session Two Alex Fadeev - Link for this PPT:
Sample Variability Consider the small population of integers {0, 2, 4, 6, 8} It is clear that the mean, μ = 4. Suppose we did not know the population mean.
One Random Variable Random Process.
A Different Type of Monte Carlo Simulation Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
Discrete Distribution Functions Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1.
EXTENSION OF LATIN HYPERCUBE SAMPLES WITH CORRELATED VARIABLES C. J. SALLABERRY, a J. C. HELTON b – S. C. HORA c aSandia National Laboratories, New Mexico.
Limits to Statistical Theory Bootstrap analysis ESM April 2006.
5-1 ANSYS, Inc. Proprietary © 2009 ANSYS, Inc. All rights reserved. May 28, 2009 Inventory # Chapter 5 Six Sigma.
ETM 607 – Random-Variate Generation
Exam 2: Rules Section 2.1 Bring a cheat sheet. One page 2 sides. Bring a calculator. Bring your book to use the tables in the back.
MONTE CARLO ANALYSIS When a system contains elements that exhibit chance in their behavior, the Monte Carlo method of simulation may be applied.
Components are existing in ONE of TWO STATES: 1 WORKING STATE with probability R 0 FAILURE STATE with probability F R+F = 1 RELIABLEFAILURE F R Selecting.
Variance Reduction Fall 2012
NESSUS Overview and General Capabilities
- 1 - Computer model under uncertainty In previous lecture on accuracy assessment –We considered mostly deterministic models. – We did not distinguish.
Introduction Paul J. Hurtado Mathematical Biosciences Institute (MBI), The Ohio State University 19 May 2014 (Monday a.m.)
Section 8-6 Testing a Claim about a Standard Deviation or Variance.
More on Monte Carlo simulation in SR GEOL/CE/EEB 8601 Intro to Stream Restoration.
MATH Section 3.1.
Effect of Recyling Events on MET results use a toy Monte Carlo to try to understand the effect that the recycling of min bias “pileup” events has on the.
EE201C Final Project Adeel Mazhar Charwak Apte. Problem Statement Need to consider reading and writing failure – Pick design point which minimizes likelihood.
MATH Section 7.5.
MATH Section 7.2.
Unified Adaptivity Optimization of Clock and Logic Signals Shiyan Hu and Jiang Hu Dept of Electrical and Computer Engineering Texas A&M University.
Computer Simulation Henry C. Co Technology and Operations Management,
Logarithmic Functions and Their Graphs
Inferences On Two Samples
Chapter 9 Hypothesis Testing
Mapped Meshing Jake Blanchard Spring 2008.
Chapter 3: Averages and Variation
Chapter 9 Hypothesis Testing
CHAPTER- 3.1 ERROR ANALYSIS.
Probability Plot Examples
Chapter 5: Graphs & Functions
Further Topics on Random Variables: Derived Distributions
MATH 2311 Section 7.5.
Distribution-free Monte Carlo for population viability analysis
Further Topics on Random Variables: Derived Distributions
RANDOM NUMBERS SET # 1:
Sampling Plans.
Presentation transcript:

Latin Hypercube Sampling Example Jake Blanchard Spring 2010 Uncertainty Analysis for Engineers1

Example Z=XY X and Y follow exponential distributions x =1 y =1/2 Uncertainty Analysis for Engineers2

Step 1 Divide cdfs into even intervals (vertical axis) Uncertainty Analysis for Engineers3

Step 2 Now sample a number in each section For example, pick a random number between 0.8 and 1.0 and use it to get a random value for x xx=expinv(rx,mux); Uncertainty Analysis for Engineers4

Step 3 Sort the values (shuffle) nux=xx(randperm(n)); Uncertainty Analysis for Engineers5

An Alternative Instead of using expinv, we can generate the inverse ourselves Just take the CDF and solve for x xx=-mux*log(1-rx); Uncertainty Analysis for Engineers6

First Script n= ; mux=1; muy=2; x=exprnd(mux,n,1); y=exprnd(muy,n,1); mz=mean(x.*y); error=abs(mz-mux*muy)/mux/muy d=linspace(0,1,n+1); rx=unifrnd(d,d+1/n); ry=unifrnd(d,d+1/n); xx=expinv(rx,mux); yy=expinv(ry,muy); nux=xx(randperm(n)); nuy=yy(randperm(n)); mz=mean(nux.*nuy); error=abs(mz-mux*muy)/mux/muy Uncertainty Analysis for Engineers7

Alternative n= ; mux=1; muy=2; x=exprnd(mux,n,1); y=exprnd(muy,n,1); mz=mean(x.*y); error=abs(mz-mux*muy)/mux/muy d=linspace(0,1,n+1); rx=unifrnd(d,d+1/n); ry=unifrnd(d,d+1/n); xx=-mux*log(1-rx); yy=-muy*log(1-ry); nux=xx(randperm(n)); nuy=yy(randperm(n)); mz=mean(nux.*nuy); error=abs(mz-mux*muy)/mux/muy Uncertainty Analysis for Engineers8

Test Z=XY X and Y are beta with mean of 1 and 2, respectively Use simple Monte Carlo Then use LHS without sorting Then use LHS with sorting N=100,000 For each case, find mean 100 times and then take standard deviation of results Uncertainty Analysis for Engineers9

Case 1 n=100000; ntrials=100; mz=zeros(ntrials,1); for i=1:ntrials x=exprnd(mux,n,1); y=exprnd(muy,n,1); mz(i)=mean(x.*y); end std(mz) Uncertainty Analysis for Engineers10

Case 2 d=linspace(0,1,n+1); for i=1:ntrials rx=unifrnd(d,d+1/n); rx=rx(1:end-1); ry=unifrnd(d,d+1/n); ry=ry(1:end-1); x=expinv(rx,mux); y=expinv(ry,muy); mz(i)=mean(x.*y); end std(mz) Uncertainty Analysis for Engineers11

Case 3 d=linspace(0,1,n+1); for i=1:ntrials rx=unifrnd(d,d+1/n); rx=rx(1:end-1); ry=unifrnd(d,d+1/n); ry=ry(1:end-1); x=expinv(rx,mux); y=expinv(ry,muy); nux=x(randperm(n)); nuy=y(randperm(n)); mz(i)=mean(nux.*nuy); end std(mz) Uncertainty Analysis for Engineers12

Results Mean(mz)Std(mz) Case Case Case Uncertainty Analysis for Engineers13