Pallavi Chitturi

Choice-Based Conjoint Analysis : Models and Designs

Introduction Conjoint Analysis (CA) Discrete Choice Experimentation (DCE) Random Utility Models The Logistic Model Contributions of the Book Some Statistical Concepts Principles of Experimental Design Experimental versus Treatment Design Balanced Incomplete Block Designs and 3-Designs Factorial Experiments Fractional Factorial Experiments Hadamard Matrices and Orthogonal Arrays Foldover Designs Mixture Experiments Estimation Transformations of the Multinomial Distribution Testing Linear Hypotheses Generic Designs Introduction Four Linear Models Used in CA and DCE Brands-Only Designs Attribute-Only Designs Brands-Plus-Attributes Designs Brands, Attributes, and Interaction Design Estimation and Hypothesis Testing Appendix: Logit Analysis of Traditional Conjoint Rating Scale Data Designs with Ordered Attributes Introduction Linear, Quadratic, and Cubic Effects Interaction Components: Linear and Quadratic An Illustration Pareto Optimal Designs Inferences on Main Effects Inferences on Main Effects in 2m Experiments Inferences on Interactions Orthogonal Polynomials Substitution Rate of Attributes Reducing Choice Set Sizes Introduction Subsetting Choice Sets Subsetting Levels into Overlapping Sets Subsetting Attributes into Overlapping Sets Designs Generated from a BIBD Cyclic Construction: s Choice Sets of Size s Each for an ss Experiment Estimating a Subset of Interactions Availability (Cross-Effects) Designs Introduction Brands-Only Availability Designs Portfolio Designs Brand and One (or More) Attributes Brands and More Than One Attribute Sequential Methods Introduction Sequential Experiment to Estimate All Two- and Three-Attribute Interactions Sequential Methods to Estimate Main Effects and Interactions, Including a Common Attribute in 2m Experiments CA Testing Main Effects and a Two-Factor Interaction Sequentially Interim Analysis Some Sequential Plans for 3m Experiments Mixture Designs Introduction Mixture Designs: CA Example Mixture Designs: DCE Example Mixture-Amount Designs Other Mixture Designs Mixture Designs: Field Study Illustration References Index

Optionen der B-to-B-Markenpositionierung

Wissenschaft und Praxis diskutieren vielfaltige Strategien, um den wahrgenommenen Wert von Produkten fur den Kunden zu erhohen und dem Preiskampf zu entgehen. In diesem Beitrag wird ein Ansatz vorgestellt, der die verschiedenen Moglichkeiten zum Aufbau von Wert durch Marken fur den Kunden (im Folgenden als Markenwert bezeichnet) im B-to-B-Bereich strukturiert und vergleicht. Dies umfasst Entscheidungen uber die Quelle des Markenwertes und die Ebene des Markeninvestments. Die Hauptquellen von Markenwert sind Differenzierung und Beziehungsaufbau. Die Hauptebenen fur das Markeninvestment sind die Produkt- und die Unternehmensebene. Die relative Vorteilhaftigkeit der verschiedenen Optionen aus Kundensicht wird mithilfe einer Reihe von Wahlexperimenten mit Einkaufsmanagern getestet. Die Analyse zeigt, dass Differenzierung wichtiger als Beziehungsaufbau ist. Ferner belegen die Ergebnisse, dass in Situationen mit geringen Kaufvolumina und geringem Risiko eher die Produktebene wichtig ist. Hingegen wird in Situationen mit hohen Preisen und hohem Risiko eher eine Kombination von Produkt-und Unternehmensebene praferiert. Diese Erkenntnisse helfen B-to-B-Unternehmen, systematischer uber die Positionierung der Marke und deren Strategie sowie die Markenrelevanz nachzudenken.

Optionen der B-to-B-Markenpositionierung – Generierung von Wert für den Kunden durch Marken

Wissenschaft und Praxis diskutieren vielfaltige Strategien, um den wahrgenommenen Wert von Produkten fur den Kunden zu erhohen und dem Preiskampf zu entgehen. In diesem Beitrag wird ein Ansatz vorgestellt, der die verschiedenen Moglichkeiten zum Aufbau von Wert durch Marken fur den Kunden (im Folgenden als Markenwert bezeichnet) im B‐to‐B‐Bereich strukturiert und vergleicht. Dies umfasst Entscheidungen uber die Quelle des Markenwertes und die Ebene des Markeninvestments. Die Hauptquellen von Markenwert sind Differenzierung und Beziehungsaufbau. Die Hauptebenen fur das Markeninvestment sind die Produkt‑ und die Unternehmensebene. Die relative Vorteilhaftigkeit der verschiedenen Optionen aus Kundensicht wird mithilfe einer Reihe von Wahlexperimenten mit Einkaufsmanagern getestet. Die Analyse zeigt, dass Differenzierung wichtiger als Beziehungsaufbau ist. Ferner belegen die Ergebnisse, dass in Situationen mit geringen Kaufvolumina und geringem Risiko eher die Produktebene wichtig ist. Hingegen wird in Situationen mit hohen Preisen und hohem Risiko eher eine Kombination von Produkt‑ und Unternehmensebene praferiert. Diese Erkenntnisse helfen B‐to‐B‐Unternehmen, systematischer uber die Positionierung der Marke und deren Strategie sowie die Markenrelevanz nachzudenken.

Design and Analysis of Experiments in the Health Sciences, First Edition, by Gerald Van Belle and Kathleen F. Kerr, John Wiley & Sons, 2012

There exist a plethora of books dealing with the design and analysis of experiments with applications in engineering and agriculture that can be used in an experimental design course for first year graduate students or in an advanced course for undergraduates. A popular book in this area is Montgomery (2012), an exhaustive text on the design and analysis of experiments. The book Design and Analysis of Experiments in the Health Sciences focuses instead on applications in health sciences. The book incorporates randomized clinical trials and microarrays into the broader context of design, and several examples in the book involve animal and human experiments in health sciences. The book is intended for readers who have had at least a first course in statistics and are familiar with basic concepts such as t-tests, simple linear regression, and hypothesis testing. It assumes knowledge of type I error, type II error, and power. These prerequisites are relatively modest, so the book can be used in an advanced statistics course for undergraduate students. The material can be covered more rapidly in a graduate-level course and can also be used as the basis of a short course for professionals in the health sciences. Chapter 1 covers the basic principles of design and analysis and includes topics such as sample size calculation, assessing assumptions, confirmatory and exploratory analysis, and missing data. There is a brief paragraph on statistical software that essentially states “most statistical computing packages will do.” The authors state that they primarily used Stata and R, and computer output in the form of tables is included with many of the examples in subsequent chapters. However, the text does not include any details for the R and Stata packages, which could be a disadvantage for readers. Chapters 2–6 discuss five basic designs in detail—completely randomized, randomized block (including Latin squares and incomplete blocks), factorial, multilevel, and repeated-measures designs. Chapter 2 introduces the completely randomized design. Topics discussed include sample size considerations and estimation and analysis for the basic design, as well as for the analysis of covariance. Chapter 3 covers the randomized block design and extensions—the Latin squares and incomplete block designs. Balanced incomplete block designs and partially balanced incomplete block designs are also mentioned. Chapter 4 discusses factorial designs with a brief mention of fractional factorials. Although factorial designs are a rich family of designs with wide applications in engineering and science, the chapter does not provide an in-depth treatment of this important family of designs. The authors point out that factorial designs are used more extensively in the industry than in health sciences due to the smaller number

A Confidence Region for Zero-Gradient Solutions for Robust Parameter Design Experiments

One of the key issues in robust parameter design is to configure the controllable factors to minimize the variance due to noise variables. However, it can sometimes happen that the number of control variables is greater than the number of noise variables. When this occurs, two important situations arise. One is that the variance due to noise variables can be brought down to zero The second is that multiple optimal control variable settings become available to the experimenter. A simultaneous confidence region for such a locus of points not only provides a region of uncertainty about such a solution, but also provides a statistical test of whether or not such points lie within the region of experimentation or a feasible region of operation. However, this situation requires a confidence region for the multiple-solution factor levels that provides proper simultaneous coverage. This requirement has not been previously recognized in the literature. In the case where the number of control variables is greater than the number of noise variables, we show how to construct critical values needed to maintain the simultaneous coverage rate. Two examples are provided as a demonstration of the practical need to adjust the critical values for simultaneous coverage.