Monte Carlo Analysis A Technique for Combining Distributions.

Slides:



Advertisements
Similar presentations
Design of Experiments Lecture I
Advertisements

Chapter 18 If mathematical analysis is too difficult, we can try each possibility out on paper. That way we can find which alternative appears to work.
Structural reliability analysis with probability- boxes Hao Zhang School of Civil Engineering, University of Sydney, NSW 2006, Australia Michael Beer Institute.
SE503 Advanced Project Management Dr. Ahmed Sameh, Ph.D. Professor, CS & IS Project Uncertainty Management.
CHAPTER 21 Inferential Statistical Analysis. Understanding probability The idea of probability is central to inferential statistics. It means the chance.
Risk Management Jan Röman OM Technology Securities Systems AB.
Chapter 10: Sampling and Sampling Distributions
Computing the Posterior Probability The posterior probability distribution contains the complete information concerning the parameters, but need often.
FIN 685: Risk Management Topic 5: Simulation Larry Schrenk, Instructor.
Copyright © 2004 David M. Hassenzahl Monte Carlo Analysis David M. Hassenzahl.
Determining the Size of
Point and Confidence Interval Estimation of a Population Proportion, p
Probability and Sampling Theory and the Financial Bootstrap Tools (Part 2) IEF 217a: Lecture 2.b Fall 2002 Jorion chapter 4.
Analysis of Variance Chapter 3Design & Analysis of Experiments 7E 2009 Montgomery 1.
Chapter 14 Simulation. Monte Carlo Process Statistical Analysis of Simulation Results Verification of the Simulation Model Computer Simulation with Excel.
The Calibration Process
Today Concepts underlying inferential statistics
BCOR 1020 Business Statistics
Understanding sample survey data
Uncertainty in Engineering - Introduction Jake Blanchard Fall 2010 Uncertainty Analysis for Engineers1.
Monté Carlo Simulation MGS 3100 – Chapter 9. Simulation Defined A computer-based model used to run experiments on a real system.  Typically done on a.
Decision analysis and Risk Management course in Kuopio
1 D r a f t Life Cycle Assessment A product-oriented method for sustainability analysis UNEP LCA Training Kit Module k – Uncertainty in LCA.
Introduction to ModelingMonte Carlo Simulation Expensive Not always practical Time consuming Impossible for all situations Can be complex Cons Pros Experience.
Monte Carlo Simulation in Decision Making
Lecture 11 Implementation Issues – Part 2. Monte Carlo Simulation An alternative approach to valuing embedded options is simulation Underlying model “simulates”
Stress testing and Extreme Value Theory By A V Vedpuriswar September 12, 2009.
Introduction to Monte Carlo Methods D.J.C. Mackay.
Math 116 Chapter 12.
Forecasting Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill.
Marko Tainio, marko.tainio[at]thl.fi Modeling and Monte Carlo simulation Marko Tainio Decision analysis and Risk Management course in Kuopio
Mote Carlo Method for Uncertainty The objective is to introduce a simple (almost trivial) example so that you can Perform.
CHAPTER 16: Inference in Practice. Chapter 16 Concepts 2  Conditions for Inference in Practice  Cautions About Confidence Intervals  Cautions About.
Analysis and Visualization Approaches to Assess UDU Capability Presented at MBSW May 2015 Jeff Hofer, Adam Rauk 1.
Biostatistics IV An introduction to bootstrap. 2 Getting something from nothing? In Rudolph Erich Raspe's tale, Baron Munchausen had, in one of his many.
Monte Carlo Simulation and Personal Finance Jacob Foley.
Simulation Prepared by Amani Salah AL-Saigaly Supervised by Dr. Sana’a Wafa Al-Sayegh University of Palestine.
LECTURE 22 VAR 1. Methods of calculating VAR (Cont.) Correlation method is conceptually simple and easy to apply; it only requires the mean returns and.
1 1 Slide © 2007 Thomson South-Western. All Rights Reserved OPIM 303-Lecture #9 Jose M. Cruz Assistant Professor.
LECTURE 19 THURSDAY, 14 April STA 291 Spring
Section 8.1 Estimating  When  is Known In this section, we develop techniques for estimating the population mean μ using sample data. We assume that.
Stat 13, Tue 5/8/ Collect HW Central limit theorem. 3. CLT for 0-1 events. 4. Examples. 5.  versus  /√n. 6. Assumptions. Read ch. 5 and 6.
Introduction to Modeling Introduction Management Models Simulate business activities and decisions Feedback about and forecast of outcomes Minimal risk.
Contemporary Engineering Economics, 6 th edition Park Copyright © 2016 by Pearson Education, Inc. All Rights Reserved Risk Simulation Lecture No. 40 Chapter.
1-1 Copyright © 2015, 2010, 2007 Pearson Education, Inc. Chapter 10, Slide 1 Chapter 10 Understanding Randomness.
The Common Shock Model for Correlations Between Lines of Insurance
Copyright © 2007 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Chapter 11 Understanding Randomness.
1 SMU EMIS 7364 NTU TO-570-N Inferences About Process Quality Updated: 2/3/04 Statistical Quality Control Dr. Jerrell T. Stracener, SAE Fellow.
1 Chapter 19 Monte Carlo Valuation. 2 Simulation of future stock prices and using these simulated prices to compute the discounted expected payoff of.
Copyright © 2012 Pearson Education. All rights reserved © 2010 Pearson Education Copyright © 2012 Pearson Education. All rights reserved. Chapter.
Estimation: Confidence Intervals Based in part on Chapter 6 General Business 704.
LECTURE 25 THURSDAY, 19 NOVEMBER STA291 Fall
Slide Understanding Randomness.  What is it about chance outcomes being random that makes random selection seem fair? Two things:  Nobody can.
1 3. M ODELING U NCERTAINTY IN C ONSTRUCTION Objective: To develop an understanding of the impact of uncertainty on the performance of a project, and to.
5-1 ANSYS, Inc. Proprietary © 2009 ANSYS, Inc. All rights reserved. May 28, 2009 Inventory # Chapter 5 Six Sigma.
FIN 614: Financial Management Larry Schrenk, Instructor.
Chapter 7 Sampling Distributions. Sampling Distribution of the Mean Inferential statistics –conclusions about population Distributions –if you examined.
Learning Simio Chapter 10 Analyzing Input Data
Chapter 19 Monte Carlo Valuation. Copyright © 2006 Pearson Addison-Wesley. All rights reserved Monte Carlo Valuation Simulation of future stock.
Copyright © 2007 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Slide
Chapter 10 Understanding Randomness. Why Be Random? What is it about chance outcomes being random that makes random selection seem fair? Two things: –
Control Charting: Lessons Learned Along the Way Why, What, and How Stacie Hammack Florida Department of Agriculture and Consumer Services Food Laboratories.
1 Life Cycle Assessment A product-oriented method for sustainability analysis UNEP LCA Training Kit Module k – Uncertainty in LCA.
Computer Simulation Henry C. Co Technology and Operations Management,
Monte Carlo Simulation
Risk Mgt and the use of derivatives
Understanding Randomness
Monte Carlo Simulation Managing uncertainty in complex environments.
Professor S K Dubey,VSM Amity School of Business
Presentation transcript:

Monte Carlo Analysis A Technique for Combining Distributions

FIN 591: Financial Modeling, Spring Purpose of lecture Introduce Monte Carlo Analysis as a tool for managing uncertainty To demonstrate how it can be used in the policy setting To discuss its uses and shortcomings, and how they are relevant to policy making processes.

FIN 591: Financial Modeling, Spring What is Monte Carlo Analysis? It is a tool for combining distributions, and thereby propagating more than just summary statistics It uses a random number generation, rather than analytic calculations It is increasingly popular due to high speed personal computers.

FIN 591: Financial Modeling, Spring Background/History “Monte Carlo” from the gambling town of the same name (no surprise) Limited use because time consuming Much more common since late 80’s Too easy now?

FIN 591: Financial Modeling, Spring Why do Monte Carlo Analysis? Combining distributions With more than two distributions, solving analytically is very difficult Simple calculations lose information Mean  mean = mean 95% %ile  95%ile  95%ile! Gets “worse” with 3 or more distributions.

FIN 591: Financial Modeling, Spring Monte Carlo Analysis Takes an equation Example: Risk = probability  consequence Draws randomly from defined distributions Multiplies, stores Repeats this over and over and over… Results displayed as a new, combined distribution.

FIN 591: Financial Modeling, Spring Simple Example Skin cream additive is an irritant Many samples of cream provide information on concentration: mean 0.02 mg chemical/application standard dev mg chemical/application Two tests show probability of irritation given application low p(effect per mg exposure)=0.05 / mg high p(effect per mg exposure)=0.10 / mg.

FIN 591: Financial Modeling, Spring Skin cream additive data PotencyExposure Information type {low, high}Mean, deviation Data{0.05, 0.10}0.02 mg, mg Distribution?Uniform? Triangular? Normal? Lognormal?

FIN 591: Financial Modeling, Spring Analytical Results Risk = Exposure  potency Mean risk = 0.02 mg  / mg = or 0.15% probability that someone using the cream will be irritated.

FIN 591: Financial Modeling, Spring Analytical results “Conservative estimate” Use upper 95 th %ile Risk = 0.03 mg  / mg = or p(irritation|application) = 0.29%.

FIN 591: Financial Modeling, Spring Monte Carlo: Visual example Exposure = normal (mean 0.02 mg, s.d. = mg) Potency = uniform (range 0.05 / mg to 0.10 / mg)

FIN 591: Financial Modeling, Spring Random Draw One p(irritate) = mg × / mg =

FIN 591: Financial Modeling, Spring Random Draw Two p(irritate) = mg × / mg = Summary: {0.0010, }

FIN 591: Financial Modeling, Spring Random Draw Three p(irritate) = mg × / mg = Summary: {0.0010, , }

FIN 591: Financial Modeling, Spring Random Draw Four p(irritate) = mg × / mg = Summary: {0.0010, , , }

FIN 591: Financial Modeling, Spring After Ten Random Draws Summary {0.0010, , , , , , , , , } Mean = Standard deviation = ( ).

FIN 591: Financial Modeling, Spring Using software Could write this program using a random number generator But, several software packages exist I User friendly Customizable RNG good up to about 10,000 iterations.

FIN 591: Financial Modeling, Spring iterations (less than two seconds) Monte Carlo results Mean Standard Deviation Compare to analytical results Mean standard deviationn/a.

FIN 591: Financial Modeling, Spring Summary chart trials

FIN 591: Financial Modeling, Spring Summary - 10,000 Trials Monte Carlo results Mean Standard Deviation Compare to analytical results Mean standard deviationn/a.

FIN 591: Financial Modeling, Spring Summary chart - 10,000 trials

FIN 591: Financial Modeling, Spring Issues: Sensitivity Analysis Which input distributions have the greatest effect on the eventual distribution Which parameters can both be influenced by policy and reduce risks When better data can be most valuable (information isn’t free…nor even cheap).

FIN 591: Financial Modeling, Spring Issues: Correlation Two distributions are correlated when a change in one is associated with a change in another Example: People who eat lots of peas may eat less broccoli (or may eat more…) Usually doesn’t have much effect unless significant correlation (|  |>0.75).

FIN 591: Financial Modeling, Spring Generating Distributions Invalid distributions create invalid results, which leads to inappropriate policies Two options Empirical Theoretical.

FIN 591: Financial Modeling, Spring Empirical Distributions Most appropriate when developed for the issue at hand. Example: local fish consumption Survey individuals or otherwise estimate Data from individuals elsewhere may be very misleading A number of very large data sets have been developed and published.

FIN 591: Financial Modeling, Spring Empirical Distributions Challenge: when there’s very little data Example of two data points Uniform distribution? Triangular distribution? Not a hypothetical issue…is an ongoing debate in the literature Key is to state clearly your assumptions Better yet…do it both ways!

FIN 591: Financial Modeling, Spring Which Distribution?

FIN 591: Financial Modeling, Spring Random number generation Shouldn’t be an is good to at least 10,000 iterations 10,000 iterations is typically enough, even with many input distributions.

FIN 591: Financial Modeling, Spring Theoretical Distributions Appropriate when there’s some mechanistic or probabilistic basis Example: small sample (say 50 test animals) establishes a binomial distribution Lognormal distributions show up often in nature, particular economics/business.

FIN 591: Financial Modeling, Spring Some Caveats Beware believing that you’ve really “understood” uncertainty Central tendencies are NOT “real risk” Distributions are only PART of uncertainty Beware misapplication Ignorance at best Fraudulent at worst.

FIN 591: Financial Modeling, Spring Example (after Finkel 1995) Alar “versus” aflatoxin Exposure has two elements Peanut butter consumption aflatoxin residue Juice consumption Alar/UDMH residue Potency has one element aflatoxin potencyUDMH potency Risk = (consumption  residue  potency)/body weight

FIN 591: Financial Modeling, Spring Inputs for Alar & aflatoxin

FIN 591: Financial Modeling, Spring Alar and Aflatoxin Point Estimates Aflatoxin estimates: Mean = Alar (UDMH) estimates: Mean =

FIN 591: Financial Modeling, Spring Alar and Aflatoxin Monte Carlo 10,000 runs Generate distributions (don’t allow 0) Don’t expect correlation.

FIN 591: Financial Modeling, Spring Aflatoxin and Alar Monte Carlo Results (Point Values)

FIN 591: Financial Modeling, Spring

FIN 591: Financial Modeling, Spring

FIN 591: Financial Modeling, Spring

FIN 591: Financial Modeling, Spring

FIN 591: Financial Modeling, Spring End