1 Regions of rationality: Maps for bounded agents (Decision Analysis, in press) Robin M. Hogarth ICREA & Universitat Pompeu Fabra, Barcelona & Natalia.

Slides:



Advertisements
Similar presentations
BUYER BEHAVIOUR INDIVIDUAL DECISION MAKING
Advertisements

Ch2 Data Preprocessing part3 Dr. Bernard Chen Ph.D. University of Central Arkansas Fall 2009.
Automated Regression Modeling Descriptive vs. Predictive Regression Models Four common automated modeling procedures Forward Modeling Backward Modeling.
Multi‑Criteria Decision Making
Hypothesis Testing Steps in Hypothesis Testing:
Combining Test Data MANA 4328 Dr. Jeanne Michalski
Simple Regression Equation Multiple Regression y = a + bx Test Score Slope y-intercept Predicted Score  y = a + b x + b x + b x ….. Predicted Score 
CHAPTER SIXTEEN Alternative Evaluation and Selection McGraw-Hill/Irwin Copyright © 2004 by The McGraw-Hill Companies, Inc. All rights reserved.
Certainty Equivalent and Stochastic Preferences June 2006 FUR 2006, Rome Pavlo Blavatskyy Wolfgang Köhler IEW, University of Zürich.
Decision making and economics. Economic theories Economic theories provide normative standards Expected value Expected utility Specialized branches like.
Visual Recognition Tutorial
Evaluating Search Engine
Chapter 7: Statistical Applications in Traffic Engineering
Chapter 7(7b): Statistical Applications in Traffic Engineering Chapter objectives: By the end of these chapters the student will be able to (We spend 3.
Bayesian Learning Rong Jin. Outline MAP learning vs. ML learning Minimum description length principle Bayes optimal classifier Bagging.
1/55 EF 507 QUANTITATIVE METHODS FOR ECONOMICS AND FINANCE FALL 2008 Chapter 10 Hypothesis Testing.
Simple Regression correlation vs. prediction research prediction and relationship strength interpreting regression formulas –quantitative vs. binary predictor.
Comparison and Combination of Ear and Face Images in Appearance-Based Biometrics IEEE Trans on PAMI, VOL. 25, NO.9, 2003 Kyong Chang, Kevin W. Bowyer,
Probability Population:
Impact Evaluation Session VII Sampling and Power Jishnu Das November 2006.
Chapter 6 Normal Probability Distributions
Lecture II-2: Probability Review
Ensemble Learning (2), Tree and Forest
INVESTOR BEHAVIOUR AND BENCHMARKS Presentation to Finansmarkedsfondet Executive Board Sari Carp Norwegian School of Management (BI) 8 December 2005.
Chapter 15 Nonparametric Statistics
Chapter 12 Inferential Statistics Gay, Mills, and Airasian
Scales and Indices While trying to capture the complexity of a phenomenon We try to seek multiple indicators, regardless of the methodology we use: Qualitative.
ANCOVA Lecture 9 Andrew Ainsworth. What is ANCOVA?
Chapter 10 Hypothesis Testing
Estimation and Hypothesis Testing. The Investment Decision What would you like to know? What will be the return on my investment? Not possible PDF for.
Chapter 6 The Normal Probability Distribution
01/24/05© 2005 University of Wisconsin Last Time Raytracing and PBRT Structure Radiometric quantities.
Statistical Decision Making. Almost all problems in statistics can be formulated as a problem of making a decision. That is given some data observed from.
Slide 1 Estimating Performance Below the National Level Applying Simulation Methods to TIMSS Fourth Annual IES Research Conference Dan Sherman, Ph.D. American.
Educational Research: Competencies for Analysis and Application, 9 th edition. Gay, Mills, & Airasian © 2009 Pearson Education, Inc. All rights reserved.
1 Statistical Distribution Fitting Dr. Jason Merrick.
What are the Odds? Sampling Theory and Logic. Let’s Be Realistic… It’s unlikely you’ll be in a position to do much sampling in your daily work Important.
Exploration Strategies for Learned Probabilities in Smart Terrain Dr. John R. Sullins Youngstown State University.
Chapter 6 USING PROBABILITY TO MAKE DECISIONS ABOUT DATA.
Managerial Economics Demand Estimation & Forecasting.
Combining multiple learners Usman Roshan. Bagging Randomly sample training data Determine classifier C i on sampled data Goto step 1 and repeat m times.
Decision Making How do people make decisions? Are there differences between making simple decisions vs. complex ones?
Bounded Rationality: The Role of Psychological Heuristics in OR Konstantinos Katsikopoulos Max Planck Institute for Human Development Center for Adaptive.
Educational Research Chapter 13 Inferential Statistics Gay, Mills, and Airasian 10 th Edition.
I271B The t distribution and the independent sample t-test.
INTRODUCTION TO Machine Learning 3rd Edition
Application of Class Discovery and Class Prediction Methods to Microarray Data Kellie J. Archer, Ph.D. Assistant Professor Department of Biostatistics.
The Order of Acquisition of Durable Goods and The Multidimensional Measurement of Poverty Joseph Deutsch and Jacques Silber August 2005 Department of Economics,
Linear Methods for Classification Based on Chapter 4 of Hastie, Tibshirani, and Friedman David Madigan.
ETHEM ALPAYDIN © The MIT Press, Lecture Slides for.
Linear Judgment Models: What Do They Suggest About Human Judgment?
T-test for dependent Samples (ak.a., Paired samples t-test, Correlated Groups Design, Within-Subjects Design, Repeated Measures, ……..) Next week: Read.
Classification Ensemble Methods 1
Regression. We have talked about regression problems before, as the problem of estimating the mapping f(x) between an independent variable x and a dependent.
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 15-1 Chapter 15 Multiple Regression Model Building Basic Business Statistics 10 th Edition.
Feature Selction for SVMs J. Weston et al., NIPS 2000 오장민 (2000/01/04) Second reference : Mark A. Holl, Correlation-based Feature Selection for Machine.
ELEC 303 – Random Signals Lecture 17 – Hypothesis testing 2 Dr. Farinaz Koushanfar ECE Dept., Rice University Nov 2, 2009.
Educational Research Inferential Statistics Chapter th Chapter 12- 8th Gay and Airasian.
Tree and Forest Classification and Regression Tree Bagging of trees Boosting trees Random Forest.
Contact Info: Improving Decision Making: The use of simple heuristics Dr. Guillermo Campitelli Cognition Research Group Edith.
Methods of multivariate analysis Ing. Jozef Palkovič, PhD.
ETHEM ALPAYDIN © The MIT Press, Lecture Slides for.
Machine Learning: Ensemble Methods
Understanding Results
Hypothesis Testing: Hypotheses
MANA 4328 Dr. Jeanne Michalski
How to handle missing data values
Statistical Methods For Engineers
The
Power and Sample Size I HAVE THE POWER!!! Boulder 2006 Benjamin Neale.
Presentation transcript:

1 Regions of rationality: Maps for bounded agents (Decision Analysis, in press) Robin M. Hogarth ICREA & Universitat Pompeu Fabra, Barcelona & Natalia Karelaia H.E.C., Université de Lausanne

2 “Regions of rationality” The starting point: –“heuristics and biases” (Kahneman, Slovic, & Tversky, 1982) –simple decision rules can rival the predictive ability of complex algorithms (e.g., regression) (e.g., TTB: Gigerenzer, Todd, & the ABC Research Group, 1999; EW : Dawes & Corrigan, 1974). Idea: –Attention as a scarce resource (Simon, 1978) -> how much information to seek & how to combine the pieces to make decisions in different “regions”: identify decision rules that are appropriate to each region multiple-cue prediction (multi-attribute choice) cues are probabilistically related to the criterion

3 A theoretical approach… 1.Effectiveness of several heuristics: the probability that the best of m alternatives (with k cues) is identified; the environmental conditions favoring various heuristics, e.g.: differential weighting of cues inter-correlations of cues continuous/binary cues (c/b) noise in the environment interactions of these factors 2. Illustration: 20 “artificial” and 4 empirical environments

4 Models Single Variable (SV) models 1.Lexicographic – SVc 2.Lexicographic – SVb 3.DEBA (binary cues) Equal weight (EW) models 4. EWc 5. EWb Hybrid models 6.EW/DEBA 7.EW/SVb Domran (DR) models (lower benchmark) 8.DRc 9.DRb Multiple regression (MR) (upper benchmark) 10.MRc 11.MRb

5 Method Single Variable, continuous cues - SVc Choosing between A & B Y = criterion and X = cue Assume: Y and X are N(0,1), >0 = error,, N(0, ), Question:

6 Prob {SVc chooses the best b/w A & B}

7 Therefore, Prob {SVc chooses the best b/w A & B} pdf = probability density function

- z1 and z2 are bivariate N Prob {SVc chooses the best from A, B, & C}

9 SVc: generalizing to the case of m alternatives (m>3) where (m-1) between-alternative comparisons

10 Overall probability of correct choice by SVc Random sampling of m=3 from the underlying population of alternatives. Either A, B, or C is chosen -> overall probability is: 3 P{((X a >X b ) & (X a >X c ))&((Y a >Y b )&(Y a >Y c ))} integrated across : where,.

11 Overall probability of correct choice by SVc: generalizing to m>3 where

12 Other models: EWc & MRc Model: Error: VdVd di*di*

13 Models with binary cues - SVb where Therefore,

14 Models with binary cues - SVb choosing 1 of 2 where

15 Models with binary cues - DEBA & Hybrids Prob {a given alternative is chosen correctly}= the joint probability that the sequence of decisions (or eliminations) made at each stage is correct. Three key notions: 1.Appropriate model for each stage 2.Partial correlations: and partial st. deviations: 3. Probability theory to calculate sequence of correct eliminations

Illustration: 20 “artificial” environments -Choosing the best from 2, 3, and 4 alternatives -n=40 k

17 Low inter-cue corrHigh inter-cue corr 3 cues 5 cues High inter-cue corrLow inter-cue corr Choosing the best from 3

18 (1) Similarity of models’ performance –agreement between models (average between all pairs, A-D)=63% (vs. 33.(3)% of random agreement), lower when lower inter-cue corr. (2)Model with continuous cues outperform their binary counterparts (except DR). –DRb > DRc. Choosing at random: DRb = in 51%, DRc = in 81%. (3)Larger inter-cue correlation reduces performance of all models (except SV). Some results

Regression of model performance

20 Illustration: 4 empirical datasets 1) Golf all-around ranking, N=60 1. Birdie average (*-1) 2. Scoring average 3. Putting average 2) Golf earnings, N=60 1. Top 10 finishes 2. All-around ranking (*-1) 3. Consecutive cuts 3) PhD economics programs: ratings-1993, N=107http:// 1. # of PhDs for the academic year to Total # of program citations 88-92/ number program faculty 3. % Faculty with research support 4) Consumer reports:test score for digital cameras, 1. Image quality 2. Picture download time 3. Focusing

21 Illustration: empirical datasets

22 Golf ranking Golf earnings Economics PhD programs Consumer reports

23 (1)Our contributions –Analytical analysis –Regions of rationality: a multidimensional terrain (2)Further research & implications –Non-random sampling of alternatives –Hybrids with categorical & continuous variables –Different loss functions –Predicting consumer preferences –Bounded rationality and expertise: how do people build maps of their decision making terrain? Discussion