Discrete Choice Analysis with R | springerprofessional.de

Springer Professional

nach oben

2022 | Buch

Kapitel lesen Erstes Kapitel lesen

Discrete Choice Analysis with R

verfasst von: Antonio Páez, Geneviève Boisjoly

Verlag: Springer International Publishing

Buchreihe : Use R!

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book is designed as a gentle introduction to the fascinating field of choice modeling and its practical implementation using the R language. Discrete choice analysis is a family of methods useful to study individual decision-making. With strong theoretical foundations in consumer behavior, discrete choice models are used in the analysis of health policy, transportation systems, marketing, economics, public policy, political science, urban planning, and criminology, to mention just a few fields of application. The book does not assume prior knowledge of discrete choice analysis or R, but instead strives to introduce both in an intuitive way, starting from simple concepts and progressing to more sophisticated ideas. Loaded with a wealth of examples and code, the book covers the fundamentals of data and analysis in a progressive way. Readers begin with simple data operations and the underlying theory of choice analysis and conclude by working with sophisticated models including latent class logit models, mixed logit models, and ordinal logit models with taste heterogeneity. Data visualization is emphasized to explore both the input data as well as the results of models. This book should be of interest to graduate students, faculty, and researchers conducting empirical work using individual level choice data who are approaching the field of discrete choice analysis for the first time. In addition, it should interest more advanced modelers wishing to learn about the potential of R for discrete choice analysis. By embedding the treatment of choice modeling within the R ecosystem, readers benefit from learning about the larger R family of packages for data exploration, analysis, and visualization.

Anzeige

Inhaltsverzeichnis

Frontmatter

Chapter 1. Data, Models, and Software

Abstract

Models propose a simplified representation of the reality, which is useful to develop a common ground for describing, analyzing, and understanding complex phenomena. Model building requires three things: Raw materials. Tools. Technical expertise (hopefully!). This is true whether the model is physical (for instance a sculpture), conceptual (a mental map), or statistical/mathematical (the gravity model or a regression model). In the case of a sculpture, the raw materials can be marble, wood, or clay; the tools chisels, mallet, and spatula; and the technique the mastery of the sculptor when working with the tools and the materials. Anyone can try sculpture, and most people can create sculptures.

Antonio Páez, Geneviève Boisjoly

Chapter 2. Exploratory Data Analysis

Abstract

Discrete choice modeling covers a family of techniques useful to infer decision-making processes in many disciplines, including economics, geography, transportation engineering and planning. These techniques are well represented in a variety of journals, including specialized outlets such as the Journal of Choice Modeling, and is a preferred tool in many applications due to the rich behavioral interpretation of the models. Thousands of applications of discrete choice analysis are found in the literature, which have greatly contributed to our understanding of individual behavior and implicit preferences. This includes studies on transportation mode choices, altruistic behavior, residential choices, and so on. Experienced modelers know well that estimating and interpreting a discrete choice model are only two aspects in a more extended data analysis process, one that ranges from data collection to presentation of results to inform policy and decision-making.

Antonio Páez, Geneviève Boisjoly

Chapter 3. Fundamental Concepts

Abstract

There are many kinds of models: analog (like sculptures, maquettes, scale models), conceptual (like mental maps), and mathematical/statistical models.

Antonio Páez, Geneviève Boisjoly

Chapter 4. Logit

Abstract

The concept of utility has many flaws—key among them is that it is not directly observable. If utility could be measured directly by an external observer (or analyst), behavior would seem deterministic. However, unlike Laplace’s Demon, an external observer with only human capabilities has limited knowledge of the conditions under which choices are made, if for no other reason that they cannot possibly know the frame of mind of the decision-maker at the moment when choices are made.

Antonio Páez, Geneviève Boisjoly

Chapter 5. Practical Issues in the Specification and Estimation of Discrete Choice Models

Abstract

In theory, there is no difference between theory and practice. But in practice, there is.

Antonio Páez, Geneviève Boisjoly

Chapter 6. Behavioral Insights from Choice Models

Abstract

In Chap. 5 we covered some important practical aspects around the estimation of the multinomial logit model. Many of them transfer to other kinds of discrete choice models as well. Before exploring other models we will take the opportunity, armed as we are with the practical skills to estimate the multinomial logit model, to see how discrete choice models can be used to understand preferences and to infer behavior.

Antonio Páez, Geneviève Boisjoly

Chapter 7. Non-proportional Substitution Patterns I: Generalized Extreme Value Models

Abstract

The multinomial logit model is the workhorse of discrete choice analysis. As seen in the preceding chapters, it is a model that is intuitive, and moreover, its closed analytical form makes it simple and convenient to estimate.

Antonio Páez, Geneviève Boisjoly

Chapter 8. Non-proportional Substitution Patterns II: The Probit Model

Abstract

In Chap. 7, the topic of non-proportional substitution was discussed, and a method for deriving logit models using the Generalized Extreme Value (GEV) system was presented. In particular, the nested (or hierarchical) logit model was introduced as an alternative modelling approach to alleviate the issues that emerge when a multinomial logit model is not fully specified. If there are hidden correlations, the proportional substitution patterns that result from the Independence from Irrelevant Alternatives property may be inappropriate.

Antonio Páez, Geneviève Boisjoly

Chapter 9. Dealing with Heterogeneity I: The Latent Class Logit Model

Abstract

Chapters 7 and 8 were concerned with substitution patterns, particularly those resulting from the Independence of Irrelevant Alternatives (IIA) property of the logit model.

Antonio Páez, Geneviève Boisjoly

Chapter 10. Dealing with Heterogeneity II: The Mixed Logit Model

Abstract

Chapter 9 introduced the latent class logit model, a technique useful to model taste variations in a sample. In this chapter, a variation on the theme will be introduced, namely the mixed logit model. We will see how the mixed logit model is related to the latent class logit model: the key difference is how the latent segments are conceptualized.

Antonio Páez, Geneviève Boisjoly

Chapter 11. Models for Ordinal Responses

Abstract

In the preceding chapters, our focus has been on modeling choices in situations where decision-makers are faced with a set of binomial or multinomial alternatives (e.g., heating systems, commute mode, etc.). The data sets used in the examples represent choices made by individuals or households among different alternatives that had one thing in common: they were measured in a categorical but not ordinal scale (q.v., Chap. 1). As such, the alternatives did not follow a natural order or a logical sequence: a gas central system is not “higher” or “greater” than an electric room system. Similarly, car is not “higher” or “greater” than walking or public transport. This is what we call unordered choices.

Antonio Páez, Geneviève Boisjoly

Backmatter

Titel: Discrete Choice Analysis with R
verfasst von: Antonio Páez
Geneviève Boisjoly
Copyright-Jahr: 2022
Verlag: Springer International Publishing
Electronic ISBN: 978-3-031-20719-8
Print ISBN: 978-3-031-20718-1
DOI: https://doi.org/10.1007/978-3-031-20719-8

Premium Partner