P 1. a = a vector of means of the distribution of individuals' p k = the probability of an individual choosing the kth X 1

Size: px
Start display at page:

Download "P 1. a = a vector of means of the distribution of individuals' p k = the probability of an individual choosing the kth X 1"

Transcription

1 USING SAS IML TO PERFORM HIERARCIDCAL BAYES ESTIMATION FOR DISCRETE CHOICE MODELING David Steenhard, LexisNexis, Louisville, KY Nan-Ting Chou, University of Louisville, Louisville, KY ABSTRACT Most methods for analyzing choice-based conjoint data do so by combining data for all individuals. There could be extreme weakness in analyzing data this way, for it could obscure important individual aspects of the data. Hierarchical Bayes Estimation is one method of estimating individual part-worths. This method can reasonably estimate individual part-worths even with relatively little data from each respondent. In this paper we provide an introduction to Hierarchical Bayes Estimation and use this algorithm written in SAS IML to perform a choice-based conjoint aualysis. Sufficient details are provided to allow the readers to easily use Hierarchical Bayes Estimation as a tool for discrete choice modeling. INTRODUCTION Disaggregate or individual discrete choice modeling is fast becoming a favorite topic among market research professionals due to the technique's ability to answer a wide range of marketing questions. Discrete choice analysis consists of a series of questions, which ask respondents to choose between two or more hypothetical products. The products are described in each question by a list of attributes, which enable the respondent to easily compare between alternative products. Discrete choice modeling is typically done by using aggregate models (models that assume that all respondents have the same preferences). The resulting analysis provides a model of the choice behavior of a representative, or average respondent. However, respondents tend to differ across socio-demographic and/or attitudinal characteristics. The difficulty with an aggregate model is that by assuming all respondents have the same preferences, we assume away the important differences among respondents, which can potentially lead to problems with the model's accuracy and a loss of valuable marketing information. Important recent advances in statistics, however, have provided a way around the data deficiency of aggregate models. Hierarchical Bayes Estimation is one method of estimating individual partworths. Landmark articles by Allen by and Ginter ( 1995) and Link, DeSarbo, Green, and Young (1996) describe the estimation of individual partworths using Hierarchical Bayes (HB) models. This approach seemed extremely promising, since it could estimate reasonable individual part-worths even with relatively little amount of data from each respondent. However, this method is computationally intensive, and usually requires many thousands of iterations before convergence is met This paper illustrates how Hierarchical Bayes Estimation in SAS IML can be used to solve a discrete choice modeling problem and outlines the mechanics of Hierarchical Bayes Estimation in detail. IDERARCHICAL BAYES MODEL The HB model used here is called "hierarchical" because it has two levels. At the upper level, we assume that individuals' part-worths are described by a mnltivariate normal distribution with the following notation. where: p,- N(a,D) P 1 =a vector ofpart-worths for the ith individual. a = a vector of means of the distribution of individuals' part-worths. D = a matrix of variances and covariances of the distribution of part-worths across individuals. At the lower level we assume that, given an individual's part-worths, his/her probabilities of choosing particular alternatives are governed by a multinomiallogit model. The probability of the ith individual choosing the kth alternative in a particular task is: where: Pt = exp( x, 'P,)IL 1 exp(x/p,) (I) p k = the probability of an individual choosing the kth concept in a particular choice task. X 1 = a vector of values describing the jth alternative in that choice task. The parameters to be estimated are the vectors P 1 partworths for each individual, the vector a means ofthe distribution of part-worths, and the matrix D the variances and co variances of that distribution. ESTIMATION OF THE PARAMETERS The parameters P, a, and D are estimated by an itemtive process. This process is quite robust, and its results do not depend on starting values. However; to make the process converge as quickly as possible, one should start with estimates of the parameters that are reasonably close to fmal values. Initial estimates of the P 1 were set equal to the parameters of the aggregate multinomiallogit model. Initial estimates of a is the avemge of the P 1 Our initial estimates of D consists of variance and covariances of the aggregate multinomiallogit model. 146

2 Given these initial values, each iteration consists of the following three steps. Using the estimates of the fl 1 and D, generate a new estimate of a. assuming that a is distributed nonnally with mean equal to the average. Using the estimates of the a and fl 1, generate a new estimate of D, from the inverse Wishart distribution. Using the estimates of the a and D, generate a new estimate of fl 1, from a procedure known as "Metropolis Hasting Algorithm" Which will be discuss in detail in the next section. For each of these three steps we re-estimate one set of the parameters (a, fl 1 or D) based on current values of the other two sets. This technique is know as "Gibbs sampling", and converges to the correct distribution for each of the three sets of parameters. Another name for this procedure is a "Monte Carlo Markov Chain", because the fact that the estimates in each iteration are determined from those of the previous iteration by a constant set of transition rules. This process is carried out for a large number of iterations. The first few thousand are used to achieve convergence, with successive iterations fitting the data better and better. These iterations are called "burn-in" or "transitory" iterations. After the transitory iterations are completed we start to save the estimates of the fl 1,a, and D for each iteration. To get a point estimates of the part-worths for each respondent, we take the average of the fl 1 from these iterations. METROPOLIS HASTINGS ALGORITHM The Metropolis Hasting Algorithm is used to draw each new set of betas for each individual. We use the symbol flold to indicate the previous estimate of fl 1 We then generate a trial value for the new estimate of fl 1, which we call fl NEW, and then test whether it represents an improvement. If so we accept it as our next estimate, if not we accept or reject it with probability depending on how much worse it is than the previous estimate. To get flnew we draw a random vector d of"differences" from a distribution with mean of zero and covariance matrix proportioual to D, and let flnew = flow + d We then calculate the probability of the data given each set ofpartworths, fl OLD and fl NEW, using the formula for the multinomial logit model (1). That is done by calculating the probability of each choice that individual made, using the multinomiallogit formula for Pk and then multiplying all these probabilities together and call these resulting values p OLD and p NEW, respectively. Next we calculate the relative density of the distribution of the betas corresponding to fl OLD and fl NEW, given current estimates of parameters a and D (these serve as priors in the Bayesian updating). Call these values dow and d NEW. The relative density of the distribution at the location of a point fl is given by the following fonnula: Relative Density=exrt-Yz (fl -a)'d- 1 (fl -a)j Finally calculate the ratio: r = PNEWdNEW I Powdow From Bayesian updating the posterior probabilities are proportional to the product of the likelihood times the priors. The probabilities p NEW and pow are the likelihood's of the data given the parameter estimates flow and fl NEW. The densities dow and d NEW are proportional to the probabilities of drawing those values of flow and flnew, respectively, from the distribution of part-worths, and play the role of priors. Therefore, r is the ratio of posterior probabilities of flow and flnew. If r is greater than unity, the new estimate has a higher posterior probability than the previous one, and we accept fl NEW. If r is less than unity we accept fl NEW with probability equal tor. Two influences are at work in deciding whether to accept the new estimate ofbeta. First, if a respondent's choices fit well, their estimated fl 1 depends mostly on his own data and is influenced less by the population distribution (relative density). But if their choices fit poorly then their estimated fl 1 depends more on the population distribution and is influenced less by their data. In this way HB makes use of every respondent's data in producing estimates for each individual. This sharing of information is what gives HB the ability to produce reasonable estimates for each respondent even when there may be inadequate information for each individual. The following SAS IML code performs the Metropolis Hasting Algorithm. start beta(nind,subj,set,x,beta,alpha,d, jd,umean,arate); /* Help find the vector delb by finding the factor that is proportional to the covariance matrix */ ucov=jd*d; accept=o; decline=o; invd=inv(d); seed=int(ranuni(o)*loooo); * Break all the information by individual respondents; do i=l to nind; xi=x[loc(subj=i),]; seti=set[loc(subj=i),]; /* To get the new estimate for beta draw a random vector delb from a multivariate normal distribution with mean of zero and covariance matrix proportional to D, 147

3 and let the new betan = beta + delb, where beta is the previous estimate of betan *I call vnormal(delb,umean,ucov,l,seed); delb=delb'; seed=seed+i; betao=beta [, i I ; betan=betao+delb; * Find the exponential of the utilities for the new and old estimates; eutilo=exp(xi*betao); eutiln=exp(xi*betan); I* Find the probability of each choice and for each choice task that the individual made then multiply all the probabilities together *I maxseti=max(seti); po=l; pn=l; do j=l to maxseti; tutilo=eutilo[loc(seti=j),]; tutiln=eutiln[loc(seti=j),); I* Find the sum of all of the exponential utilities for each choice task *I sutilo=sum(tutilo); sutiln=sum(tutiln); I* Find the probability of each choice that the individual made then calculate the product of the probabilities for each individual *I ptempn= tutilnlsutiln; ptempo= tutilolsutilo; do m=l to nrow(ptempn); po=po*ptempo[m,); pn=pn*ptempn[m,); I* Calculate the relative density of the distribution of the betas corresponding to betao and betan given current estimates of parameters alpha and D. Call these values etmpo and etmpn. Finally calculate the ratio (pn*etmpn)/po*etmpo) *I diffo=betao-alpha; diffn=betan-alpha; tmpo=diffo'*invd*diffo; tmpn=diffn'*invd*diffn; etmpo=exp(-o.s*tmpo); etmpn=exp(-o.s*tmpn); I* Select either betao or betan for the new estimate of beta based on the ratio. If this ratio is greater than or equal to unity accept betan as the new estimate for beta for that individual. If the ratio less than unity, then use a random process to decide whether to accept betan or retain betao. Accept betan with probability equal to the ratio. *I ratio=(pn*etmpn)l(po*etmpo); minr=min(ratio,l); rand=uniform(o); I* Determine if you want to save the new estimate of beta or not *I if rand <= minr then do; beta[,i)=betan; accept=accept+l; else do; beta[,i)=betao; decline=decline + 1; arate=accept/(accept+decline); free xi seti delb eutilo eutiln maxseti po pn tutilo tutiln sutilo sutiln ptempn ptempo invd diffo diffn tmpo tmpn etmpo etmpn ratio minr rand accept decline; finish; DISCRETE CHOICE MODEL EXAMPLE A discrete choice model was developed to find the perceptions of 530 customers that conduct company research. These customers have either used a fee-based product within the past six months, or would find such a product of value. In order to qualify for this study, respondents had to be at least somewhat involved in the decision to purchase fee-based research product for their company. The survey was developed to present respondents with different trade-offs of attnbutes, such as: features and functions, content, price plans, and brands. Survey respondents were asked to evaluate descriptions of a set of attributes and indicate which one they would most likely buy. The feature levels and prices for each alternative are systematically varied. Each respondent evaluate sixteen different ''buying situations" or choice sets. Subsequent analysis of the choices made by each individual allows us to estimate the importance of the various levels of each attribute of the product or service offering. A key deliverable from discrete choice modeling is a market simulator that allows managers to continue to test "what if' scenarios by varying the attributes such as price, features and bundling services. In this study, ten different attributes were included in the trade-off design (see appendix A for a description of the attributes). Each attribute has at least two different levels. Combinations of attribute levels that were not feasible were omitted. Two scenarios were simulated for this analysis. The first simulation was with "Basic" Brand A and its three major competitors: Brands B, C, and D. This simulation was designed to represent the product offering available in the current marketplace. The second simulation was with "Deluxe Brand A" and its three major competitors. "Deluxe Brand A" includes: upgrades of screening capabilities and downloading capabilities; increased coverage of geographic area; and raising the subscription prices. The competition's product offerings remained the same for both simulations. 148

4 RESULTS Using the Hierarchical Bayes Estimation in SAS IML to analyze the response data from the 530 customers, we are able to derive the following results: Estimates of the preference shares for the different brands in the current market. This provides information about customers' preferences of the existing products. Estimates of the preference shares for the "Deluxe Brand A" product. It allows us to predict the change of customers' preference towards Brand A if the deluxe version is offered to the market. Run different "What If' scenarios by varying the different attributes and measuring the preference share it has on the different brands. Estimates of individual and overall average (across all individuals) utility values for each level of every attribute. This allows us to predict the potential improvement of preference share for each attribute of the existing product. Estimates of average (across all individuals) importance of each attribute. This provides information about which attributes are more important in the produces composition. The following table displays importance of an attribute. The importance is defined as its weight, or maximum influence it can have on a product choice, given the range of attribute levels defined in the study. Attribute importances are only accurate when individual or segment level utility data are available. Attribute Average Importance Price Plan 12.6% Brand 13.1% Screening Capabilities 4.6% Supplemental Content 9.0"/o Download/Editing Capabilities 9.7% Company Coverage 14.6% Updates oftime-sensitive lnfonnation 4.1% Linking to Full-Text Source Documents 5.0"/o Content Selection 3.1% Subscription Price 9.7% Transaction Price 5.8% Report Price 8.7% 100.0"/o From the importances table attributes price plan, brand, and company coverage are the top 3 attributes. CONCLUSIONS Discrete choice modeling has several advantages that make it appropriate for use in market research. It offers a more realistic shopping simulation-respondents are asked to choose between carefully chosen sets of complete products. They show the trade-offs the respondents make between attributes/features. This is much more realistic than the most common forms of ratings-based or preference based conjoint analysis exercises. Choice-based conjoint analysis allow participants to indicate to indicate that they would not purchase any of the alternatives. It is possible in choice-based conjoint analysis to have "product or alternative specific" attribute levels. For example in this paper we use different price plan, but the attributes describing the price plans are different among the different pricing plans. Hierarchical Bayes also provide individual-level estimates from choice data which have been shown to out-perform aggregate logit for predicting shares for holdout choices, even when there was very little heterogeneity in the data. Having individual-level utilities lends to segmented marketing strategies. REFERENCES Sawtooth Software (2000). The CBC/HB Module, For Hierarchical Bayes Estimation, Version 1.5, Sequim, W A:. Sawtooth Software. Sawtooth Software (1999). The Client Conjoint Simulator, version 1.0, Sequim, W A: Sawtooth Software. Allenby, G.M. and Ginter, J.L. (1995). ''Using Extremes to Design Products and Segment Markets," Journal of Marketing Research, 32, (November) Lenk, P.J., DeSarbo, W.S., Green P.E. and Young, M.R. (1996) "Hierarchical Bayes Conjoint Analysis: Recovery of Part-Worth Heterogeneity fro Reduced Experimental Designs," Marketing Science, 15, Chib, S. and Greenberg, E. (1995) ''Understanding the Metropolis-Hasting Algorithm," American Statistician, 49, (November) SAS Institute Inc. (1999). SASIIML User's Guide, Version 8, Cary NC: SAS Institute Inc. CONTACT INFORMATION Your comments and questions are valued and encouraged. Contact the authors at: David Steenhard Hidden Pond Ct. Louisville, KY david steenhard@lexis-nexis.com Nan-Ting Chou University of Louisville Economics Department University of Louisville Louisville, KY ntchouo SAS is a registered trademark ofsas Institute Inc. in the USA and other countries. indicates USA registration. 149

5 DISCRETE CHOICE DESIGN ATTRIBUTES (APPENDIX A) Price Plan: Flat rate by # of users Transactional per report Flat rate by# of committed reports Brand: Brand A Brand B Brand C Brand D Screening Capabilities: Sophisticated Limited Supplemental Content: Limited Moderate Extensive Extensive plus legal, patent & trademark Download/Editing Capabilities: Print only, no download Download, no edit Download and edit Company Coverage: U.S. public companies only All U.S. (public & private companies) All U.S. plus U.K. and Europe All U.S. plus U.K., Europe, Asia, & Latin America Subscription Price: Subscription Price Plan 1 Subscription Price Plan 2 Subscription Price Plan 3 Subscription Price Plan 4 Transaction Price: Price 1 Price 2 Price 3 Reports Price: Report Price Plan 1 Report Price Plan 2 Report Price Plan 3 Report Price Plan 4 Updates of Time-Sensitive Information: Real time Recent time Linking to Full-Text Source Documents: Cannot link Can link Content Selection: One part at a time Parts/entire report, in one step 150

THREE LEVEL HIERARCHICAL BAYESIAN ESTIMATION IN CONJOINT PROCESS

THREE LEVEL HIERARCHICAL BAYESIAN ESTIMATION IN CONJOINT PROCESS Please cite this article as: Paweł Kopciuszewski, Three level hierarchical Bayesian estimation in conjoint process, Scientific Research of the Institute of Mathematics and Computer Science, 2006, Volume

More information

Benefit-Risk Assessment Using Bayesian Choice-Based Conjoint: An Example

Benefit-Risk Assessment Using Bayesian Choice-Based Conjoint: An Example Benefit-Risk Assessment Using Bayesian Choice-Based Conjoint: An Example Kimberley Dilley Panel Session: Bayesian Methods in Assessing Benefit-Risk Preference in a Structured Framework Report on work performed

More information

Sawtooth Software Conference

Sawtooth Software Conference Sawtooth Software Conference Glossary of Terms Welcome to the Sawtooth Software Conference! If you are relatively new to conjoint analysis and to this conference, you may find that some speakers tend to

More information

An Introduction to Choice-Based Conjoint

An Introduction to Choice-Based Conjoint An Introduction to Choice-Based Conjoint with Sawtooth Software s Lighthouse Studio 2 Part 2 ANALYSIS & MARKET SIMULATIONS 3 Our CBC Exercise Brand Style Color Options* Price Samsung Side by Side Stainless

More information

Sawtooth Software. Comparing Hierarchical Bayes Draws and Randomized First Choice for Conjoint Simulations RESEARCH PAPER SERIES

Sawtooth Software. Comparing Hierarchical Bayes Draws and Randomized First Choice for Conjoint Simulations RESEARCH PAPER SERIES Sawtooth Software RESEARCH PAPER SERIES Comparing Hierarchical Bayes Draws and Randomized First Choice for Conjoint Simulations Bryan K. Orme and Gary C. Baker, Sawtooth Software, Inc., 2000 Copyright

More information

Real World Performance of Choice-Based Conjoint Models. Martin Natter Markus Feurstein

Real World Performance of Choice-Based Conjoint Models. Martin Natter Markus Feurstein Real World Performance of Choice-Based Conjoint Models Martin Natter Markus Feurstein Report No. 58 June 2001 June 2001 SFB Adaptive Information Systems and Modelling in Economics and Management Science

More information

Sawtooth Software. Which Conjoint Method Should I Use? RESEARCH PAPER SERIES. Bryan K. Orme Sawtooth Software, Inc.

Sawtooth Software. Which Conjoint Method Should I Use? RESEARCH PAPER SERIES. Bryan K. Orme Sawtooth Software, Inc. Sawtooth Software RESEARCH PAPER SERIES Which Conjoint Method Should I Use? Bryan K. Orme Sawtooth Software, Inc. Copyright 2013, Sawtooth Software, Inc. 1457 E 840 N Orem, UT 0 84097 (801) 477-4700 www.sawtoothsoftware.com

More information

Learning User Real-Time Intent for Optimal Dynamic Webpage Transformation

Learning User Real-Time Intent for Optimal Dynamic Webpage Transformation Learning User Real-Time Intent for Optimal Dynamic Webpage Transformation Amy Wenxuan Ding, Shibo Li and Patrali Chatterjee Web Appendix: Additional Results for the Two-State Proposed Model A. Program

More information

Managerial Overview of Conjoint Analysis

Managerial Overview of Conjoint Analysis Chapter 1 Managerial Overview of Conjoint Analysis A great deal of market research commissioned today is descriptive in nature rather than predictive. Descriptive information is useful to characterize

More information

Conjoint Analysis: Marketing Engineering Technical Note 1

Conjoint Analysis: Marketing Engineering Technical Note 1 Conjoint Analysis: Marketing Engineering Technical Note 1 Table of Contents Introduction Conjoint Analysis for Product Design Designing a conjoint study Using conjoint data for market simulations Transforming

More information

What Is Conjoint Analysis? DSC 410/510 Multivariate Statistical Methods. How Is Conjoint Analysis Done? Empirical Example

What Is Conjoint Analysis? DSC 410/510 Multivariate Statistical Methods. How Is Conjoint Analysis Done? Empirical Example What Is Conjoint Analysis? DSC 410/510 Multivariate Statistical Methods Conjoint Analysis 1 A technique for understanding how respondents develop preferences for products or services Also known as trade-off

More information

Individual Level or Segmentation Based Market Simulation? Martin Natter Markus Feurstein

Individual Level or Segmentation Based Market Simulation? Martin Natter Markus Feurstein Individual Level or Segmentation Based Market Simulation? Martin Natter Markus Feurstein Working Paper No. 41 June 1999 June 1999 SFB Adaptive Information Systems and Modelling in Economics and Management

More information

Sawtooth Software. Sample Size Issues for Conjoint Analysis Studies RESEARCH PAPER SERIES. Bryan Orme, Sawtooth Software, Inc.

Sawtooth Software. Sample Size Issues for Conjoint Analysis Studies RESEARCH PAPER SERIES. Bryan Orme, Sawtooth Software, Inc. Sawtooth Software RESEARCH PAPER SERIES Sample Size Issues for Conjoint Analysis Studies Bryan Orme, Sawtooth Software, Inc. 1998 Copyright 1998-2001, Sawtooth Software, Inc. 530 W. Fir St. Sequim, WA

More information

Sawtooth Software. Brief Managerial Overview of Conjoint (Trade-Off) Analysis RESEARCH PAPER SERIES. Bryan Orme, Sawtooth Software, Inc.

Sawtooth Software. Brief Managerial Overview of Conjoint (Trade-Off) Analysis RESEARCH PAPER SERIES. Bryan Orme, Sawtooth Software, Inc. Sawtooth Software RESEARCH PAPER SERIES Brief Managerial Overview of Conjoint (Trade-Off) Analysis Bryan Orme, Sawtooth Software, Inc. Copyright 2003, Sawtooth Software, Inc. 530 W. Fir St. Sequim, WA

More information

Online Technical Appendix: Key Word. Aggregations. Adaptive questions. Automobile primary and secondary. Automobiles. markets

Online Technical Appendix: Key Word. Aggregations. Adaptive questions. Automobile primary and secondary. Automobiles. markets Online Technical Appendix: Key Word Aggregations Original Word Adaptive question design Adaptive questions Adoption process Advertising and media research Advertising response functions Advertising response

More information

Sawtooth Software. The CBC System for Choice-Based Conjoint Analysis. Version 9 TECHNICAL PAPER SERIES. Sawtooth Software, Inc.

Sawtooth Software. The CBC System for Choice-Based Conjoint Analysis. Version 9 TECHNICAL PAPER SERIES. Sawtooth Software, Inc. Sawtooth Software TECHNICAL PAPER SERIES The CBC System for Choice-Based Conjoint Analysis Version 9 Sawtooth Software, Inc. 1 Copyright 1993-2017, Sawtooth Software, Inc. 1457 E 840 N Orem, Utah +1 801

More information

Applied Conjoint Analysis

Applied Conjoint Analysis Vithala R. Rao Applied Conjoint Analysis < ) Springer 1 Problem Setting 1 1.1 Introduction 1 1.1.1 Marketing Decisions and Role of Consumer Choice... 1 1.1.2 A Framework for Understanding Consumer Choice

More information

Choosing Optimal Designs for Pair-Wise Metric Conjoint Experiments

Choosing Optimal Designs for Pair-Wise Metric Conjoint Experiments Seoul Journal of Business Volume 10, Number 1 (June 2004) Choosing Optimal Designs for Pair-Wise Metric Conjoint Experiments Jin Gyo Kim* Sloan School of Management MIT, MA USA Abstract This paper presents

More information

Sawtooth Software. Using Utility Constraints to Improve the Predictability of Conjoint Analysis RESEARCH PAPER SERIES

Sawtooth Software. Using Utility Constraints to Improve the Predictability of Conjoint Analysis RESEARCH PAPER SERIES Sawtooth Software RESEARCH PAPER SERIES Using Utility Constraints to Improve the Predictability of Conjoint Analysis Bryan K. Orme and Richard M. Johnson, Sawtooth Software, 1997 Copyright 1997-2002, Sawtooth

More information

Discrete choice analysis has been a

Discrete choice analysis has been a arketing Research Forum Disaggregate Discrete Choice Assuming consumers are not alike enhances a marketer's decision-making ability. By Tm Renken Tim Renken is Vice Pn;sident, Senior Methodologist, Angus

More information

Bayesian Estimation of Random-Coefficients Choice Models Using Aggregate Data

Bayesian Estimation of Random-Coefficients Choice Models Using Aggregate Data University of Pennsylvania ScholarlyCommons Statistics Papers Wharton Faculty Research 4-2009 Bayesian Estimation of Random-Coefficients Choice Models Using Aggregate Data Andrés Musalem Duke University

More information

SIMULATING VARIABILITY OF REWORK COST AND MARKET PERFORMANCE ESTIMATES IN PRODUCT REDESIGN

SIMULATING VARIABILITY OF REWORK COST AND MARKET PERFORMANCE ESTIMATES IN PRODUCT REDESIGN Proceedings of the ASME 2015 International Design Engineering Technical Conferences and Computes and Information in Engineering Conference IDETC/CIE 2015 August 2-5, 2015, Boston, Massachusetts, USA DETC2015-47598

More information

Sawtooth Software. The ACA/Web v6.0 Technical Paper TECHNICAL PAPER SERIES

Sawtooth Software. The ACA/Web v6.0 Technical Paper TECHNICAL PAPER SERIES Sawtooth Software TECHNICAL PAPER SERIES The ACA/Web v6.0 Technical Paper Copyright 1991-2007, Sawtooth Software, Inc. 530 W. Fir St. Sequim, WA 98382 (360) 681-2300 www.sawtoothsoftware.com ACA System

More information

The Need for Market Segmentation in Buy-Till-You-Defect Models

The Need for Market Segmentation in Buy-Till-You-Defect Models Need for market segmentation 1 of 46 - April 24, 2014 The Need for Market Segmentation in Buy-Till-You-Defect Models E. Korkmaz, D. Fok, R. Kuik Buy-till-you-defect [BTYD] models are built for companies

More information

Economic Valuation of Product Features and Patent Applications

Economic Valuation of Product Features and Patent Applications Economic Valuation of Product Features and Patent Applications May 2014 Greg Allenby (OSU), Jeff Brazell (The Modelers), John Howell (PSU), Peter Rossi (UCLA) Economic Valuation of Product Features If

More information

THE CONCEPT OF CONJOINT ANALYSIS Scott M. Smith, Ph.D.

THE CONCEPT OF CONJOINT ANALYSIS Scott M. Smith, Ph.D. THE CONCEPT OF CONJOINT ANALYSIS Scott M. Smith, Ph.D. Marketing managers are faced with numerous difficult tasks directed at assessing future profitability, sales, and market share for new product entries

More information

Choice Modeling Analytics Benefits Of New Methods

Choice Modeling Analytics Benefits Of New Methods Choice Modeling Analytics Benefits Of New Methods By John V. Colias Over the past two decades, the marketing research industry has witnessed rapid growth in the use of choice models. These types of models

More information

Optimal Product Design by Sequential Experiments in High Dimensions

Optimal Product Design by Sequential Experiments in High Dimensions Optimal Product Design by Sequential Experiments in High Dimensions by Mingyu Joo Ohio State University joo.85@osu.edu Michael L. Thompson The Procter and Gamble Company thompson.ml@pg.com Greg M. Allenby

More information

Conjoint Analysis : Data Quality Control

Conjoint Analysis : Data Quality Control University of Pennsylvania ScholarlyCommons Wharton Research Scholars Wharton School April 2005 Conjoint Analysis : Data Quality Control Peggy Choi University of Pennsylvania Follow this and additional

More information

We develop a structural demand model that endogenously captures the effect of out-of-stocks on customer

We develop a structural demand model that endogenously captures the effect of out-of-stocks on customer MANAGEMENT SCIENCE Vol. 56, No. 7, July 2010, pp. 1180 1197 issn 0025-1909 eissn 1526-5501 10 5607 1180 informs doi 10.1287/mnsc.1100.1170 2010 INFORMS Structural Estimation of the Effect of Out-of-Stocks

More information

A Method for Handling a Large Number of Attributes in Full Profile Trade-Off Studies

A Method for Handling a Large Number of Attributes in Full Profile Trade-Off Studies A Method for Handling a Large Number of Attributes in Full Profile Trade-Off Often product developers need to evaluate a large number of product features, measure some interaction terms, e.g., brand and

More information

Consumer Demand Preference Patterns:

Consumer Demand Preference Patterns: Consumer Demand Preference Patterns: Using PROC IML* to Forecast Future Demand Daric Brummett, University of Notre Dame Lawrence Marsh, University of Notre Dame Economic models of consumer demand generally

More information

Generative Models for Networks and Applications to E-Commerce

Generative Models for Networks and Applications to E-Commerce Generative Models for Networks and Applications to E-Commerce Patrick J. Wolfe (with David C. Parkes and R. Kang-Xing Jin) Division of Engineering and Applied Sciences Department of Statistics Harvard

More information

Inference and computing with decomposable graphs

Inference and computing with decomposable graphs Inference and computing with decomposable graphs Peter Green 1 Alun Thomas 2 1 School of Mathematics University of Bristol 2 Genetic Epidemiology University of Utah 6 September 2011 / Bayes 250 Green/Thomas

More information

Marketing Research to Support the Stage Gate New Product Development Process

Marketing Research to Support the Stage Gate New Product Development Process Marketing Research to Support the Stage Gate New Product Development Process Agile. Iterative. Informed. Fast. These are hallmarks of today s effective new product development, with the goal of establishing

More information

Bayesian Statistics and Marketing

Bayesian Statistics and Marketing Bayesian Statistics and Marketing WILEY SERIES IN PROBABILITY AND STATISTICS Established by WALTER A. SHEWHART and SAMUEL S. WILKS Editors: DavidJ.Balding,PeterBloomfield,NoelA.C.Cressie, Nicholas I. Fisher,

More information

Sawtooth Software. ACA, CBC or Both? Effective Strategies for Conjoint Research RESEARCH PAPER SERIES. Bryan K. Orme, Sawtooth Software 1997

Sawtooth Software. ACA, CBC or Both? Effective Strategies for Conjoint Research RESEARCH PAPER SERIES. Bryan K. Orme, Sawtooth Software 1997 Sawtooth Software RESEARCH PAPER SERIES ACA, CBC or Both? Effective Strategies for Conjoint Research Bryan K. Orme, Sawtooth Software 1997 Copyright 1997-2002, Sawtooth Software, Inc. 530 W. Fir St. Sequim,

More information

Who s got the coupon? Estimating Consumer Preferences. and Coupon Usage from Aggregate Information. Andrés Musalem Eric T. Bradlow Jagmohan S.

Who s got the coupon? Estimating Consumer Preferences. and Coupon Usage from Aggregate Information. Andrés Musalem Eric T. Bradlow Jagmohan S. Who s got the coupon? Estimating Consumer Preferences and Coupon Usage from Aggregate Information Andrés Musalem Eric T. Bradlow Jagmohan S. Raju October 2007 Andrés Musalem is Assistant Professor of Marketing

More information

Managing a Successful CBC Project

Managing a Successful CBC Project Managing a Successful CBC Project - from the Conjoint Experts 2 Agenda Proposal Uncover Business Objectives & Listen for Trigger Words Sample & Survey Design The Conjoint Exercise Attributes & Levels Design

More information

Who s got the coupon? Estimating Consumer Preferences. and Coupon Usage from Aggregate Information. Andres Musalem Eric T. Bradlow Jagmohan S.

Who s got the coupon? Estimating Consumer Preferences. and Coupon Usage from Aggregate Information. Andres Musalem Eric T. Bradlow Jagmohan S. Who s got the coupon? Estimating Consumer Preferences and Coupon Usage from Aggregate Information Andres Musalem Eric T. Bradlow Jagmohan S. Raju April 2005 Andres Musalem is a doctoral candidate, The

More information

Learning Structured Preferences

Learning Structured Preferences Learning Structured Preferences Leon Bergen 1, Owain R. Evans, Joshua B. Tenenbaum {bergen, owain, jbt}@mit.edu Department of Brain and Cognitive Sciences Massachusetts Institute of Technology Cambridge,

More information

An Introduction to Choice-Based Conjoint

An Introduction to Choice-Based Conjoint An Introduction to Choice-Based Conjoint with Sawtooth Software s Lighthouse Studio 2 Agenda What is Conjoint Analysis? What is Choice Based Conjoint Analysis? Creating a CBC Exercise in Lighthouse Studio

More information

Conjoint Case. - Forte Hotel Design - MKTG 490 Prof. Conor Henderson. February 22, 2014 by Stefan Doeblitz

Conjoint Case. - Forte Hotel Design - MKTG 490 Prof. Conor Henderson. February 22, 2014 by Stefan Doeblitz Conjoint Case - Forte Hotel Design - MKTG 490 Prof. Conor Henderson February 22, 2014 by Stefan Doeblitz New Hotel Chain Fort Inn Executives, a U.K. hotel chain, plans to enter the U.S. market with a new

More information

GLMs the Good, the Bad, and the Ugly Ratemaking and Product Management Seminar March Christopher Cooksey, FCAS, MAAA EagleEye Analytics

GLMs the Good, the Bad, and the Ugly Ratemaking and Product Management Seminar March Christopher Cooksey, FCAS, MAAA EagleEye Analytics Antitrust Notice The Casualty Actuarial Society is committed to adhering strictly to the letter and spirit of the antitrust laws. Seminars conducted under the auspices of the CAS are designed solely to

More information

How Much Does the Far Future Matter? A Hierarchical Bayesian Analysis of the Public s Willingness to Mitigate Ecological Impacts of Climate Change

How Much Does the Far Future Matter? A Hierarchical Bayesian Analysis of the Public s Willingness to Mitigate Ecological Impacts of Climate Change How Much Does the Far Future Matter? A Hierarchical Bayesian Analysis of the Public s Willingness to Mitigate Ecological Impacts of Climate Change David F. Layton University of Washington Richard A. Levine

More information

Sawtooth Software. Dealing with Product Similarity in Conjoint Simulations RESEARCH PAPER SERIES

Sawtooth Software. Dealing with Product Similarity in Conjoint Simulations RESEARCH PAPER SERIES Sawtooth Software RESEARCH PAPER SERIES Dealing with Product Similarity in Conjoint Simulations Joel Huber, Duke University Bryan K. Orme, Sawtooth Software, Inc. and Richard Miller, Consumer Pulse 1999

More information

Getting Started with OptQuest

Getting Started with OptQuest Getting Started with OptQuest What OptQuest does Futura Apartments model example Portfolio Allocation model example Defining decision variables in Crystal Ball Running OptQuest Specifying decision variable

More information

ANALYSIS OF MULTI-CATEGORY PURCHASE INCIDENCE DECISIONS USING IRI MARKET BASKET DATA

ANALYSIS OF MULTI-CATEGORY PURCHASE INCIDENCE DECISIONS USING IRI MARKET BASKET DATA ANALYSIS OF MULTI-CATEGORY PURCHASE INCIDENCE DECISIONS USING IRI MARKET BASKET DATA Siddhartha Chib, P. B. Seetharaman and Andrei Strijnev ABSTRACT Empirical studies in Marketing have typically characterized

More information

Webinar Advanced Topics in CBC Part 1

Webinar Advanced Topics in CBC Part 1 Advanced Topics in CBC Part 1 2 Agenda Lightning Review of CBC Dual Response None Constant Sum/Chip Allocation Conditional Pricing/Display Alternative Specific 3 Choice-Based Conjoint (CBC) It became popular

More information

INTRODUCTION BACKGROUND. Paper

INTRODUCTION BACKGROUND. Paper Paper 354-2008 Small Improvements Causing Substantial Savings - Forecasting Intermittent Demand Data Using SAS Forecast Server Michael Leonard, Bruce Elsheimer, Meredith John, Udo Sglavo SAS Institute

More information

The Total Economic Impact Of SAS Customer Intelligence Solutions Intelligent Advertising For Publishers

The Total Economic Impact Of SAS Customer Intelligence Solutions Intelligent Advertising For Publishers A Forrester Total Economic Impact Study Commissioned By SAS Project Director: Dean Davison February 2014 The Total Economic Impact Of SAS Customer Intelligence Solutions Intelligent Advertising For Publishers

More information

Olin Business School Master of Science in Customer Analytics (MSCA) Curriculum Academic Year. List of Courses by Semester

Olin Business School Master of Science in Customer Analytics (MSCA) Curriculum Academic Year. List of Courses by Semester Olin Business School Master of Science in Customer Analytics (MSCA) Curriculum 2017-2018 Academic Year List of Courses by Semester Foundations Courses These courses are over and above the 39 required credits.

More information

Sawtooth Software. The Importance Question in ACA: Can It Be Omitted? RESEARCH PAPER SERIES

Sawtooth Software. The Importance Question in ACA: Can It Be Omitted? RESEARCH PAPER SERIES Sawtooth Software RESEARCH PAPER SERIES The Importance Question in ACA: Can It Be Omitted? W. Christopher King, Aaron Hill and Bryan Orme, Sawtooth Software, Inc. Copyright 2005, Sawtooth Software, Inc.

More information

Big Data. Methodological issues in using Big Data for Official Statistics

Big Data. Methodological issues in using Big Data for Official Statistics Giulio Barcaroli Istat (barcarol@istat.it) Big Data Effective Processing and Analysis of Very Large and Unstructured data for Official Statistics. Methodological issues in using Big Data for Official Statistics

More information

Appendix E: Conjoint Analysis

Appendix E: Conjoint Analysis Relative importance VI Appendix E: Conjoint Analysis Price is most important for the senior consumers in this case. Followed by Golden Age channel, other channels, and movie frequency. 318 320 Cable Channel

More information

Active Learning for Conjoint Analysis

Active Learning for Conjoint Analysis Peter I. Frazier Shane G. Henderson snp32@cornell.edu pf98@cornell.edu sgh9@cornell.edu School of Operations Research and Information Engineering Cornell University November 1, 2015 Learning User s Preferences

More information

Integrating Market and Credit Risk Measures using SAS Risk Dimensions software

Integrating Market and Credit Risk Measures using SAS Risk Dimensions software Integrating Market and Credit Risk Measures using SAS Risk Dimensions software Sam Harris, SAS Institute Inc., Cary, NC Abstract Measures of market risk project the possible loss in value of a portfolio

More information

Conference Presentation

Conference Presentation Conference Presentation Bayesian Structural Equation Modeling of the WISC-IV with a Large Referred US Sample GOLAY, Philippe, et al. Abstract Numerous studies have supported exploratory and confirmatory

More information

Discriminant models for high-throughput proteomics mass spectrometer data

Discriminant models for high-throughput proteomics mass spectrometer data Proteomics 2003, 3, 1699 1703 DOI 10.1002/pmic.200300518 1699 Short Communication Parul V. Purohit David M. Rocke Center for Image Processing and Integrated Computing, University of California, Davis,

More information

INSE 6300/4/UU-Quality Assurance in Supply Chain Management (Winter 2007) Mid-Term Exam

INSE 6300/4/UU-Quality Assurance in Supply Chain Management (Winter 2007) Mid-Term Exam INSE 6300/4/UU-Quality Assurance in Supply Chain Management (Winter 2007) Mid-Term Exam Professor: J. Bentahar Date: Thursday, March 15, 2007 Duration: 120 minutes NAME: ID: INSTRUCTIONS: Answer all questions

More information

Which Smart Electricity Services Contracts will Consumers Accept?

Which Smart Electricity Services Contracts will Consumers Accept? Which Smart Electricity Services Contracts will Consumers Accept? The Demand for Compensation in a Platform Market Dr. Laura Lucia Richter NERA Economic Consulting & University of Cambridge CEEM RESEARCH

More information

Residential Peak Corps Market Study: An Application of Choice-Based Conjoint Analysis Using Hierarchical Bayes Estimation

Residential Peak Corps Market Study: An Application of Choice-Based Conjoint Analysis Using Hierarchical Bayes Estimation Residential Peak Corps Market Study: An Application of Choice-Based Conjoint Analysis Using Hierarchical Bayes Estimation Vikki Wood, Sacramento Municipal Utility District Rajan Sambandam, The Response

More information

Genomic models in bayz

Genomic models in bayz Genomic models in bayz Luc Janss, Dec 2010 In the new bayz version the genotype data is now restricted to be 2-allelic markers (SNPs), while the modeling option have been made more general. This implements

More information

Appendix A Mixed-Effects Models 1. LONGITUDINAL HIERARCHICAL LINEAR MODELS

Appendix A Mixed-Effects Models 1. LONGITUDINAL HIERARCHICAL LINEAR MODELS Appendix A Mixed-Effects Models 1. LONGITUDINAL HIERARCHICAL LINEAR MODELS Hierarchical Linear Models (HLM) provide a flexible and powerful approach when studying response effects that vary by groups.

More information

New Product Development

New Product Development Principles of Marketing Global Edition Kotler and Armstrong Chapter 9: New Product Development and Product Life-Cycle Strategies Copyright 2016 Pearson Education, Inc. 9-1 New Product Development Copyright

More information

Harbingers of Failure: Online Appendix

Harbingers of Failure: Online Appendix Harbingers of Failure: Online Appendix Eric Anderson Northwestern University Kellogg School of Management Song Lin MIT Sloan School of Management Duncan Simester MIT Sloan School of Management Catherine

More information

LadyBug a software environment for stochastic epidemic models

LadyBug a software environment for stochastic epidemic models LadyBug a software environment for stochastic epidemic models Department of Statistics University of Munich, Germany CSDA 2005 Limassol, Cyprus, 28-31 October 2005 Overview LadyBug A program for simulation

More information

Chapter Seven. New-Product Development and Product Life-Cycle Strategies

Chapter Seven. New-Product Development and Product Life-Cycle Strategies Chapter Seven New-Product Development and Product Life-Cycle Strategies New-Product Development and Product Life-Cycle Strategies Topic Outline New-Product Development Strategy New-Product Development

More information

Estimation of variance components for Nordic red cattle test-day model: Bayesian Gibbs sampler vs. Monte Carlo EM REML

Estimation of variance components for Nordic red cattle test-day model: Bayesian Gibbs sampler vs. Monte Carlo EM REML Estimation of variance components for Nordic red cattle test-day model: Bayesian Gibbs sampler vs. Monte Carlo EM REML M.H. Lidauer 1, P. Madsen 2, K. Matilainen 1, E.A. Mäntysaari 1, I. Strandén 1, R.

More information

Tutorial Segmentation and Classification

Tutorial Segmentation and Classification MARKETING ENGINEERING FOR EXCEL TUTORIAL VERSION v171025 Tutorial Segmentation and Classification Marketing Engineering for Excel is a Microsoft Excel add-in. The software runs from within Microsoft Excel

More information

Which Smart Electricity Services Contracts will Consumers Accept?

Which Smart Electricity Services Contracts will Consumers Accept? Which Smart Electricity Services Contracts will Consumers Accept? The Demand for Compensation in a Platform Market Laura Lucia Richter Faculty of Economics & EPRG University of Cambridge CEEPR/EPRG/EDF

More information

Bayesian Imputation for Anonymous Visits in CRM Data

Bayesian Imputation for Anonymous Visits in CRM Data Bayesian Imputation for Anonymous Visits in CRM Data Julie Novak IBM T.J. Watson Research Center jenovak@us.ibm.com Elea McDonnell Feit Drexel University efeit@drexel.edu Shane T. Jensen The Wharton School

More information

Computational Challenges for Real-Time Marketing with Large Datasets

Computational Challenges for Real-Time Marketing with Large Datasets Computational Challenges for Real-Time Marketing with Large Datasets Alan Montgomery Associate Professor Carnegie Mellon University Tepper School of Business e-mail: alan.montgomery@cmu.edu web: http://www.andrew.cmu.edu/user/alm3

More information

Oracle Value Chain Planning Demantra Demand Management

Oracle Value Chain Planning Demantra Demand Management Oracle Value Chain Planning Demantra Demand Management Is your company trying to be more demand driven? Do you need to increase your forecast accuracy or quickly converge on a consensus forecast to drive

More information

Optimizing appointment driven systems via IPA

Optimizing appointment driven systems via IPA Optimizing appointment driven systems via IPA with applications to health care systems BMI Paper Aschwin Parmessar VU University Amsterdam Faculty of Sciences De Boelelaan 1081a 1081 HV Amsterdam September

More information

CS 188: Artificial Intelligence Fall Announcements

CS 188: Artificial Intelligence Fall Announcements CS 188: Artificial Intelligence Fall 2007 Lecture 18: Bayes Nets III 10/30/2007 Dan Klein UC Berkeley Announcements Project shift: Project 4 moved back a little Instead, mega-mini-homework, worth 3x, graded

More information

Announcements. CS 188: Artificial Intelligence Fall Inference. Reminder: Alarm Network. Normalization Trick. Inference by Enumeration?

Announcements. CS 188: Artificial Intelligence Fall Inference. Reminder: Alarm Network. Normalization Trick. Inference by Enumeration? CS 188: Artificial Intelligence Fall 2007 Lecture 18: Bayes Nets III 10/30/2007 Announcements Project shift: Project 4 moved back a little Instead, mega-mini-homework, worth 3x, graded Contest is live

More information

A Smart Approach to Analyzing Smart Meter Data

A Smart Approach to Analyzing Smart Meter Data A Smart Approach to Analyzing Smart Meter Data Ted Helvoigt, Evergreen Economics (Lead Author) Steve Grover, Evergreen Economics John Cornwell, Evergreen Economics Sarah Monohon, Evergreen Economics ABSTRACT

More information

TDWI Analytics Principles and Practices

TDWI Analytics Principles and Practices TDWI. All rights reserved. Reproductions in whole or in part are prohibited except by written permission. DO NOT COPY Previews of TDWI course books offer an opportunity to see the quality of our material

More information

Leveraging Information Across Categories

Leveraging Information Across Categories Quantitative Marketing and Economics, 1, 425 465, 2003. # 2003 Kluwer Academic Publishers. Printed in The United States. Leveraging Information Across Categories RAGHURAM IYENGAR* Graduate School of Business,

More information

Sawtooth Software. Learning Effects in Preference Tasks: Choice-Based Versus Standard Conjoint RESEARCH PAPER SERIES

Sawtooth Software. Learning Effects in Preference Tasks: Choice-Based Versus Standard Conjoint RESEARCH PAPER SERIES Sawtooth Software RESEARCH PAPER SERIES Learning Effects in Preference Tasks: Choice-Based Versus Standard Conjoint Joel Huber, Duke University Dick R. Wittink, Cornell University Richard M. Johnson, Sawtooth

More information

Value Creation. Abstract:

Value Creation. Abstract: Value Creation Abstract: If you are a Seller anticipating a sale transaction soon or in the future, you should be concerned with how your company will be viewed in the minds of the potential Buyers and

More information

Using Analytical Marketing Optimization to Achieve Exceptional Results WHITE PAPER

Using Analytical Marketing Optimization to Achieve Exceptional Results WHITE PAPER Using Analytical Marketing Optimization to Achieve Exceptional Results WHITE PAPER SAS White Paper Table of Contents Optimization Defined... 1 Prioritization, Rules and Optimization a Method Comparison...

More information

Folia Oeconomica Stetinensia DOI: /foli FORECASTING RANDOMLY DISTRIBUTED ZERO-INFLATED TIME SERIES

Folia Oeconomica Stetinensia DOI: /foli FORECASTING RANDOMLY DISTRIBUTED ZERO-INFLATED TIME SERIES Folia Oeconomica Stetinensia DOI: 10.1515/foli-2017-0001 FORECASTING RANDOMLY DISTRIBUTED ZERO-INFLATED TIME SERIES Mariusz Doszyń, Ph.D., Associate Prof. University of Szczecin Faculty of Economics and

More information

The Sawtooth Software Market Simulator. (A Supplement to the CBC v2.6 Manual)

The Sawtooth Software Market Simulator. (A Supplement to the CBC v2.6 Manual) The Sawtooth Software Market Simulator (A Supplement to the CBC v2.6 Manual) (Updated 26 January 06) Sawtooth Software, Inc. Sequim, WA http://www.sawtoothsoftware.com In this manual, we refer to product

More information

Whitepaper. Manual vs. Automated Matches

Whitepaper. Manual vs. Automated Matches Whitepaper Manual vs. Automated Matches Product Comparison as an Essential Retail Element Whether a retailer monitors competitors to make sure its prices are optimal or uses an aggressive pricing strategy

More information

Haplotype Based Association Tests. Biostatistics 666 Lecture 10

Haplotype Based Association Tests. Biostatistics 666 Lecture 10 Haplotype Based Association Tests Biostatistics 666 Lecture 10 Last Lecture Statistical Haplotyping Methods Clark s greedy algorithm The E-M algorithm Stephens et al. coalescent-based algorithm Hypothesis

More information

Primaxis Technology Ventures Inc. Business Plan Template

Primaxis Technology Ventures Inc. Business Plan Template Primaxis Technology Ventures Inc. Business Plan Template This template outlines the specific information that Primaxis Technology Ventures looks for when making an investment decision. Please understand

More information

Kuhn-Tucker Estimation of Recreation Demand A Study of Temporal Stability

Kuhn-Tucker Estimation of Recreation Demand A Study of Temporal Stability Kuhn-Tucker Estimation of Recreation Demand A Study of Temporal Stability Subhra Bhattacharjee, Catherine L. Kling and Joseph A. Herriges Iowa State University Contact: subhra@iastate.edu Selected Paper

More information

1 Services To Be Provided

1 Services To Be Provided 1 Services To Be Provided 1.1 Business Context IV&V Terms of Reference The Madrid System, the original version of which was created in 1891, is a system permitting the international registration of trademarks

More information

Practical Application of Predictive Analytics Michael Porter

Practical Application of Predictive Analytics Michael Porter Practical Application of Predictive Analytics Michael Porter October 2013 Structure of a GLM Random Component observations Link Function combines observed factors linearly Systematic Component we solve

More information

Research Methodology

Research Methodology Research Methodology 4.0 Introduction 4.1 Statement of the Problem 4.2 Research objectives 4.3 Research Hypothesis 4.4 Motivation for the study 4.5 Research Design 4.6 Sampling Element 4.7 Sampling Design

More information

CHAPTER 3 IDENTIFICATION OF THE MOST PREFERRED ATTRIBUTE USING CONJOINT ANALYSIS

CHAPTER 3 IDENTIFICATION OF THE MOST PREFERRED ATTRIBUTE USING CONJOINT ANALYSIS CHAPTER 3 IDENTIFICATION OF THE MOST PREFERRED ATTRIBUTE USING CONJOINT ANALYSIS In a heterogeneous distributed computing environment, the weighted nodes scheduling principles assigns the number of requests

More information

Promotional strategies Do s and Don ts in promotions

Promotional strategies Do s and Don ts in promotions Promotional strategies Do s and Don ts in promotions Radha Dogra, Lecturer Neville Wadia Institute of Management Studies & Research (NWIMSR) Text: Promotion is an essential element in the marketing mix.

More information

Inspection Qualification and Implementation of ENIQ in Sweden

Inspection Qualification and Implementation of ENIQ in Sweden 4th International CANDU In-service Inspection Workshop and NDT in Canada 2012 Conference, 2012 June 18-21, Toronto, Ontario ABSTRACT Inspection Qualification and Implementation of ENIQ in Sweden Tommy

More information

A Hierarchical Bayes Model for Ranked Conjoint Data

A Hierarchical Bayes Model for Ranked Conjoint Data A Hierarchical Bayes Model for Ranked Conjoint Data Prof. Dr. Thorsten Teichert Edlira Shehu* Hamburg, June 2005 *Correspondence to: Edlira Shehu University of Hamburg, Chair for Marketing and Innovation

More information

A Critical Test of Self-Explicated Utility Measurement vs. Decompositional Conjoint

A Critical Test of Self-Explicated Utility Measurement vs. Decompositional Conjoint A Critical Test of Self-Explicated Utility Measurement vs. Decompositional Conjoint Paul Riedesel, Ph.D. action marketing research 2003 A.R.T. Forum: Monterey Many Streams of Methods Developments in discrete

More information

PUTTING YOUR SEGMENTATION WHERE IT BELONGS Implementation Tactics for Taking your Segmentation off the Bookshelf and into the Marketplace

PUTTING YOUR SEGMENTATION WHERE IT BELONGS Implementation Tactics for Taking your Segmentation off the Bookshelf and into the Marketplace Putting your Segmentation Where it Belongs 1 PUTTING YOUR SEGMENTATION WHERE IT BELONGS Implementation Tactics for Taking your Segmentation off the Bookshelf and into the Marketplace For many firms, market

More information

Testing: The Critical Success Factor in the Transition to ICD-10

Testing: The Critical Success Factor in the Transition to ICD-10 Testing: The Critical Success Factor in the Transition to ICD-10 The United States (US) adopted the International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) in 1979. During

More information

Chart your future with predictive analytics

Chart your future with predictive analytics IBM Analytics Feature Guide IBM SPSS Modeler Chart your future with predictive analytics Finding hidden trends in your data can give you tremendous insights into your business. 2 Chart Your Future Contents

More information