Leeflang / Wieringa / Bijmolt Modeling Markets

Analyzing Marketing Phenomena and Improving Marketing Decision Making
1. Auflage 2014
ISBN: 978-1-4939-2086-0
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 417 Seiten

Reihe: International Series in Quantitative Marketing

This book is about how models can be developed to represent demand and supply on markets, where the emphasis is on demand models. Its primary focus is on models that can be used by managers to support marketing decisions. The market environment is changing rapidly and constantly. Prior to the introduction of scanner equipment in retail outlets, ACNielsen, the major supplier of information on brand performance, claimed that its business was to provide the score but not to explain or predict it. With technological advances and the introduction of the Internet, the opportunity to obtain meaningful estimates of demand functions has vastly improved; models that provide information about the sensitivity of market behavior to marketing activities such as advertising, pricing, promotions and distribution are now routinely used by managers for the identification of changes in marketing programs that can improve brand performance. In today's environment of information overload, the challenge is to make sense of the data that is being provided globally, in real time, from thousands of sources. Modeling Markets presents a comprehensive overview of the tools and methodologies that managers can use in decision making. It has long been known that even simple models outperform judgments in predicting outcomes in a wide variety of contexts. More complex models potentially provide insights about structural relations not available from casual observations. Although marketing models are now widely accepted, the quality of the marketing decisions is critically dependent upon the quality of the models on which those decisions are based. In this book, the authors present a wealth of insights developed at the forefront of the field, covering all key aspects of specification, estimation, validation and use of models. The most current insights and innovations in quantitative marketing are presented, including in-depth discussion of Bayesian estimation methods. Throughout the book, the authors provide examples and illustrations. This book will be of interest to researchers, analysts, managers and students who want to understand, develop or use models of marketing phenomena.

Autoren/Hrsg.

Leeflang, Peter S. H.

Wieringa, Jaap E.

Bijmolt, Tammo H. A.

Bijmolt, T.H.A

Pauwels, Koen H.

Inhaltsverzeichnis

1;Preface;8
2;Contents;10
3;1 Building Models for Markets;16
3.1;1.1 Introduction;16
3.2;1.2 Verhouten Case;17
3.3;1.3 Typologies of Marketing Models;19
3.3.1;1.3.1 Introduction;19
3.3.2;1.3.2 Decision Models Versus Models That Advance Marketing Knowledge;19
3.3.3;1.3.3 Degree of Explicitness;22
3.3.3.1;1.3.3.1 Implicit Models;22
3.3.3.2;1.3.3.2 Verbal Models;23
3.3.3.3;1.3.3.3 Formalized Models;24
3.3.3.4;1.3.3.4 Numerically Specified Models;26
3.3.4;1.3.4 Intended Use: Descriptive, Predictive and Normative Models;28
3.3.5;1.3.5 Level of Demand;29
3.4;1.4 Benefits from Using Marketing Decision Models;30
3.4.1;1.4.1 Direct Benefits;30
3.4.2;1.4.2 Indirect Benefits;31
3.5;1.5 The Model Building Process;33
3.6;1.6 Outline;36
3.7;References;37
4;2 Model Specification;40
4.1;2.1 Introduction;40
4.2;2.2 Model Criteria;41
4.2.1;2.2.1 Implementation Criteriasubject]Implementation Criteria Related to Model Structure;41
4.2.2;2.2.2 Models Should Be Simple;41
4.2.3;2.2.3 Models Should Be Built in an Evolutionary Way;44
4.2.4;2.2.4 Models Should Be Complete on Important Issues;45
4.2.5;2.2.5 Models Should Be Adaptive;47
4.2.6;2.2.6 Models Should Be Robust;48
4.3;2.3 Model Elements;49
4.4;2.4 Specification of the Functional Form;52
4.4.1;2.4.1 Models Linear in Parameters and Variables;52
4.4.2;2.4.2 Models Linear in Parameters But Not in Variables;53
4.4.3;2.4.3 Models That Are Nonlinear in Parameters, But Linearizable;56
4.4.4;2.4.4 Models That Are Nonlinear in Parameters and Not Linearizable;58
4.5;2.5 Moderation and Mediation Effects;59
4.6;2.6 Formalized Models for the Verhouten Case;61
4.7;2.7 Including Heterogeneity;63
4.8;2.8 Marketing Dynamics;65
4.8.1;2.8.1 Introduction;65
4.8.2;2.8.2 Modeling Lagged Effects: One Explanatory Variable;67
4.8.3;2.8.3 Modeling Lagged Effects: Several Explanatory Variables;73
4.8.4;2.8.4 Lead Effects;75
4.9;References;76
5;3 Data;79
5.1;3.1 Introduction;79
5.2;3.2 Data Structures;80
5.3;3.3 ``Good Data'';80
5.3.1;3.3.1 Availability;81
5.3.2;3.3.2 Quality;82
5.3.3;3.3.3 Variability;82
5.3.4;3.3.4 Quantity;83
5.4;3.4 Data Characteristics and Model Choice;84
5.5;3.5 Data Sources;85
5.5.1;3.5.1 Introduction;85
5.5.2;3.5.2 Classification;87
5.5.3;3.5.3 Internal Data;88
5.5.4;3.5.4 External Data;89
5.5.4.1;3.5.4.1 Store (Retail) Level;90
5.5.4.2;3.5.4.2 Manufacturer Level;90
5.5.4.3;3.5.4.3 Evaluation;90
5.5.4.4;3.5.4.4 Scanner Data;91
5.5.4.5;3.5.4.5 HandScan Panels;93
5.5.4.6;3.5.4.6 Causal Data;93
5.5.4.7;3.5.4.7 Other Data Inputs;93
5.5.5;3.5.5 Household Data and/or Store Level Data?;94
5.5.6;3.5.6 Big Data;95
5.5.7;3.5.7 Subjective Data;97
5.5.7.1;3.5.7.1 Justification;97
5.5.7.2;3.5.7.2 Obtaining Subjective Estimates;98
5.6;References;106
6;4 Estimation and Testing;109
6.1;4.1 Introduction;109
6.2;4.2 The General Linear Model;110
6.2.1;4.2.1 One Explanatory Variable;110
6.2.2;4.2.2 The K-Variable Case;112
6.2.3;4.2.3 Model Assumptions;114
6.3;4.3 Statistical Inference;116
6.3.1;4.3.1 Goodness of Fit;116
6.3.2;4.3.2 Assessing Statistical Significance;120
6.4;4.4 Numerically Specified Models for the Verhouten Case;125
6.5;4.5 Estimating Pooled Models;129
6.5.1;4.5.1 Introduction;129
6.5.2;4.5.2 Estimating Unit-by-Unit Models;130
6.5.3;4.5.3 Estimating Fully Pooled Modelssubject]Fully pooled models;130
6.5.4;4.5.4 Estimating Partially Pooled Models;131
6.6;References;133
7;5 Validation and Testing;135
7.1;5.1 Introduction;135
7.2;5.2 Testing the Six Basic Assumptions of the GeneralLinear Model;136
7.2.1;5.2.1 Nonzero Expectation;138
7.2.2;5.2.2 Heteroscedasticity;140
7.2.3;5.2.3 Correlated Disturbances;143
7.2.4;5.2.4 Nonnormal Errors;147
7.2.5;5.2.5 Endogenous Predictor Variables;150
7.2.6;5.2.6 Multicollinearity;152
7.2.6.1;5.2.6.1 Solutions to Multicollinearity;154
7.3;5.3 Mediation Tests;157
7.4;5.4 Joint Tests, Pooling Tests and Causality Tests;158
7.4.1;5.4.1 Joint Tests;158
7.4.2;5.4.2 Pooling Tests;161
7.4.3;5.4.3 Causality Tests;162
7.5;5.5 Face Validity;166
7.6;5.6 Model Selection;167
7.6.1;5.6.1 Introduction;167
7.6.2;5.6.2 Nested Models;167
7.6.3;5.6.3 Non-nested Models;171
7.7;5.7 Predictive Validity;173
7.8;5.8 Model Validation for the Verhouten Case;179
7.8.1;5.8.1 Testing the Six Assumptions for the Verhouten Case;180
7.8.2;5.8.2 Assessing Predictive Validity for the Verhouten Case;184
7.9;References;185
8;6 Re-estimation: Introduction to More AdvancedEstimation Methods;189
8.1;6.1 Introduction;189
8.2;6.2 Generalized Least Squares;190
8.2.1;6.2.1 Introduction;190
8.2.2;6.2.2 GLS and Heteroscedasticity;191
8.2.3;6.2.3 GLS and Autocorrelation;193
8.2.4;6.2.4 Using Generalized Least Squares with Panel Data;194
8.3;6.3 The Verhouten Case Revisited;198
8.3.1;6.3.1 Multicollinearity;199
8.3.2;6.3.2 Autocorrelation;201
8.3.3;6.3.3 Heteroscedasticity;201
8.4;6.4 Maximum Likelihood Estimation;202
8.4.1;6.4.1 Maximizing the Likelihood;202
8.4.2;6.4.2 Large Sample Properties of the MLE;205
8.4.3;6.4.3 MLE with Explanatory Variables;207
8.4.4;6.4.4 Statistical Tests;210
8.4.5;6.4.5 MLE with Explanatory Variables: An Example;212
8.5;6.5 Simultaneous Systems of Equations;214
8.6;6.6 Instrumental Variables Estimation;219
8.7;6.7 Tests for Endogeneity;223
8.8;6.8 Bayesian Estimation;225
8.8.1;6.8.1 Subjective Data;225
8.8.2;6.8.2 Combining Objective and Subjective Data: Bayes' Theorem;227
8.8.3;6.8.3 Likelihood, Prior and Posteriors;229
8.8.4;6.8.4 Conjugate Priors;230
8.8.5;6.8.5 Markov Chain Monte Carlo (MCMC) Estimation;230
8.8.6;6.8.6 Bayesian Analysis in Marketing;231
8.8.7;6.8.7 Example: Bayesian Analysis of the SCAN*PRO Model;232
8.9;References;235
9;7 Examples of Models for Aggregate Demand;237
9.1;7.1 Introduction;237
9.2;7.2 An Introduction to Individual and Aggregate Demand;238
9.3;7.3 Example of Descriptive/Predictive Models;241
9.3.1;7.3.1 Product Class Sales Models;241
9.3.2;7.3.2 Brand Sales Models;244
9.3.2.1;7.3.2.1 Introduction;244
9.3.2.2;7.3.2.2 Modeling Brand Sales Directly: The SCAN*PRO Modelsubject]SCAN*PRO model;245
9.3.2.3;7.3.2.3 Modeling Brand Sales Directly: Models for Pharmaceutical Markets;246
9.3.2.4;7.3.2.4 Modeling Brand Sales Indirectly;250
9.3.3;7.3.3 Market Share Models;251
9.3.3.1;7.3.3.1 Attraction Models;251
9.3.3.2;7.3.3.2 Own-Brand Elasticities;252
9.3.3.3;7.3.3.3 Cross-Brand Elasticities;253
9.4;7.4 Examples of Normative/Prescriptive Models;257
9.4.1;7.4.1 Introduction and Illustrations;257
9.4.1.1;7.4.1.1 Basic Model;257
9.4.1.2;7.4.1.2 Determination of the Short-Term Advertising Budget;259
9.4.1.3;7.4.1.3 Determination of the Long-Term Advertising Budget;261
9.4.2;7.4.2 Other Normative Models;263
9.4.3;7.4.3 Allocation Models;264
9.5;Appendix: The Dorfman–Steiner Theorem;266
9.6;References;268
10;8 Individual Demand Models;274
10.1;8.1 Introduction;274
10.2;8.2 Choice Models;275
10.2.1;8.2.1 Introduction;275
10.2.2;8.2.2 Binary Choice Models Specification;277
10.2.2.1;8.2.2.1 Basic Model;277
10.2.2.2;8.2.2.2 Estimation;279
10.2.2.3;8.2.2.3 Numerical Examples;280
10.2.2.4;8.2.2.4 Validation;281
10.2.2.5;8.2.2.5 Empirical Example;282
10.2.3;8.2.3 Multinomial Choice Models;283
10.2.3.1;8.2.3.1 Structure;283
10.2.3.2;8.2.3.2 Heterogeneity;290
10.2.4;8.2.4 Markov Models;291
10.2.4.1;8.2.4.1 Markov Models;291
10.2.4.2;8.2.4.2 Hidden Markov Models;296
10.3;8.3 Purchase Quantity Models;298
10.3.1;8.3.1 General Structure;298
10.3.2;8.3.2 Heterogeneity in Count Models;299
10.4;8.4 Purchase Timing: Duration Models;301
10.4.1;8.4.1 Introduction;301
10.4.2;8.4.2 Hazard Models;302
10.4.3;8.4.3 Heterogeneity in Duration Models;304
10.4.4;8.4.4 Estimation and Validation of Duration Models;306
10.5;8.5 Integrated Models;308
10.5.1;8.5.1 Integrate Incidence, Timing and Choice;308
10.5.2;8.5.2 Tobit Models;309
10.5.2.1;8.5.2.1 Introduction;309
10.5.2.2;8.5.2.2 Type-1 Tobit Model;310
10.5.2.3;8.5.2.3 Type-2 Tobit Model;311
10.6;References;314
11;9 Examples of Database Marketing Models;319
11.1;9.1 Introduction;319
11.2;9.2 Data for Database Marketing;320
11.3;9.3 Modeling Customer Life Time Value;322
11.4;9.4 Models for Customer Selection and Acquisition;325
11.4.1;9.4.1 Models for Customer Selection;325
11.4.2;9.4.2 Models for Customer Acquisition;327
11.5;9.5 Models for Customer Development;330
11.6;9.6 Models for Customer Retention;331
11.6.1;9.6.1 Models to Support Loyalty/Reward Programmes;331
11.6.2;9.6.2 Churn Prediction Models;333
11.6.2.1;9.6.2.1 Introduction;333
11.6.2.2;9.6.2.2 Aggregation;334
11.6.2.3;9.6.2.3 Validation Criteria;334
11.6.2.4;9.6.2.4 Application;336
11.7;9.7 Models for Customer Engagement;338
11.7.1;9.7.1 Customer Engagement and Customer Management;338
11.7.2;9.7.2 Customer Engagement and Acquisition/Selection;340
11.7.3;9.7.3 Customer Engagement and Customer Development;343
11.7.4;9.7.4 Customer Engagement and Retention;344
11.8;9.8 Summary of Database Marketing Models;344
11.9;References;344
12;10 Use: Implementation Issues;349
12.1;10.1 Introduction;349
12.2;10.2 Model Related Dimensions;350
12.2.1;10.2.1 Cost–Benefit Considerationssubject]Cost-benefit considerations;350
12.2.1.1;10.2.1.1 Benefits;351
12.2.1.2;10.2.1.2 Costssubject]Model costs;351
12.2.2;10.2.2 Supply and Demand of Marketing Response Models;352
12.2.2.1;10.2.2.1 Introduction;352
12.2.2.2;10.2.2.2 Supply Side;354
12.2.2.3;10.2.2.3 Demand Side;359
12.3;10.3 Organizational Validity;361
12.3.1;10.3.1 Personal Factors;361
12.3.2;10.3.2 Interpersonal Factors: The Model User–Model Builder Interface;362
12.3.3;10.3.3 Organizational Factors;364
12.4;10.4 Implementation Strategy Dimensions;365
12.4.1;10.4.1 Introduction;365
12.4.2;10.4.2 Evolutionary Model Building;366
12.4.3;10.4.3 Model Scopesubject]Model scope;367
12.4.3.1;10.4.3.1 Global Versus Local Modelssubject]Global model;368
12.4.3.2;10.4.3.2 General Versus Detailed Descriptors of Marketing Variables;368
12.4.4;10.4.4 Ease of Use;368
12.5;10.5 Marketing Management Support Systems (MMSS), Dashboards and Metrics;369
12.5.1;10.5.1 Introduction;369
12.5.2;10.5.2 Marketing Management Support Systems (MMSS)subject]Marketing Management Support Systems (MMSS);370
12.5.3;10.5.3 Dashboardssubject]Dashboards;373
12.5.4;10.5.4 Metricssubject]Metrics;375
12.6;References;379
13;Appendix A: Matrix Algebra;384
13.1;A.1 Matrices and Simple Matrix Operations;384
13.2;A.2 Matrix Multiplication;386
13.3;A.3 Special Matrices;388
13.4;A.4 Matrix Inverse;390
13.5;A.5 Determinants;392
13.6;A.6 Eigenvalues and Eigenvectors;394
13.7;A.7 Definiteness of a Matrix;397
13.8;A.8 Matrix and Vector Differentiation;398
14;Author Index;401
15;Subject Index;412