Francesca Bassi

Estimating True Changes when Categorical Panel Data are Affected by Uncorrelated and Correlated Classification Errors An Application to Unemployment Data

Conclusions about changes in categorical characteristics based on observed panel data can be incorrect when (even a small amount of) measurement error is present. Random measurement errors, referred to as independent classification errors, usually lead to over-estimation of the total amount of gross change, whereas systematic, correlated errors usually cause underestimation of the transitions. Furthermore, the patterns of true change may be seriously distorted by independent or systematic classification errors. Latent class models and directed log-linear analysis are excellent tools to correct for both independent and correlated measurement errors. An extensive example on labor market states taken from the Survey of Income and Program Participation panel is presented.

Latent class factor models for market segmentation: an application to pharmaceuticals

Market segmentation consists in defining homogeneous groups of customers and it is one of the building blocks of effective marketing planning. Various statistical techniques are available in order to identify market segments, among these, cluster analysis and latent class models. In this paper, an extension of traditional latent class analysis, called latent class factor model, is applied to market segmentation. This approach is more parsimonious than the traditional one and allows to obtain results easier to interpret in managerial terms. The model is used to identify segments of doctors homogeneous for attitude towards pharmaceutical industries and drugs prescription behaviour.

Latent class analysis for marketing scale development

Measurement scales are a crucial instrument in marketing research for measuring unobservable variables such as attitudes, opinions and beliefs. In using, evaluating or developing multi-item scales, a number of guidelines and procedures are recommended, to ensure that the measure applied is psychometrically robust. These procedures have been outlined in the psychometric literature since the late 1970s and are composed of steps that refer to construct and domain definition, scale validity, reliability, dimensionality and generalisability. Various statistical instruments are used in the scale-developing process, almost always referring to metric variables (interval or ratio scales). Instead, items forming scales are rarely measured metrically; items are frequently ordinal and, in some rare cases, nominal. In this paper, it is shown how the implementation of latent class analysis may improve the process of measurement scale development, since it explicitly considers that items generate ordinal or even nominal variables. Specifically, applying appropriate latent class models allows us to assess scale validity and reliability more soundly than traditionally used methods.

Experiential Goods and Customer Satisfaction: An Application to Films

Abstract The aim of this paper is to develop an instrument to measure customer satisfaction with reference to the entire consumption experience of an experiential product, with specific application to cinema films. Experience is defined as a new dimension of product offer: a combination of goods and services enriched by sensations. Experiential marketing has innovative features, with effects on all phases constituting a consumption experience. The research looked for important aspects in the consumption process related to satisfaction by means of a literature review and an exploratory survey. A list of items was tested on a sample population and the scale was evaluated for validity and reliability, with satisfactory results.

Latent Class Models for Marketing Strategies An Application to the Italian Pharmaceutical Market

In this paper, an extension of the latent class (LC) approach is applied to analyse the Italian pharmaceutical market. This sector is characterised by a high level of competitiveness, more limited budgets than years ago and, at the same time, expensive sales and promotional activities; in this context, it is very important to understand which factors influence doctors in prescribing medicines, so as to design appropriate marketing strategies. A special adaptation of the multilevel LC model is estimated to identify market segments, that is, groups of doctors similar in their attitude towards the work of pharmaceutical representatives.

Dynamic Segmentation of Financial Markets: A Mixture Latent Class Markov Approach

The latent class approach is innovative and flexible and can provide suitable solutions to several problems regarding the development of marketing strategies, because it takes into account specific features of the data, such as their scale of measure (often ordinal or categorical, rather than continuous), their hierarchical structure and their longitudinal component. Dynamic segmentation is of key importance in many markets where it is unrealistic to assume stationary segments due to the dynamics in consumers’ needs and product choices. In this paper, a mixture latent class Markov model is proposed to dynamically segment Italian households with reference to financial products ownership.

Evaluating quality of the didactics at university: the opportunities offered by latent class modeling

Purpose Students’ evaluation of teaching quality plays a major role in higher education. Satisfaction is not directly observable, nevertheless it can be measured through multi-item measurement scales. These instruments are extremely useful and their importance requires accurate development and validation procedures. The purpose of this paper is to show how latent class (LC) analysis can improve the procedures for developing and validating a multi-item measurement scale for measuring students’ evaluation of teaching and, at the same time, provide a deeper insight in the phenomenon under investigation. Design/methodology/approach The traditional literature highlights specific protocols along with the statistical instruments to be used for achieving this goal. However, these tools are suited for metric variables but they are adopted even when the nature of the observed variables is different, as it often occurs, since in many cases the items are ordinal. LC analysis takes explicitly into account the ordinal nature of the variables and also the fact that the object of interest is unobservable. Findings The data refer to the questionnaire to evaluate didactics to the students of the University of Padua. Within LC analysis allows an insight of scale properties, such as dimensionality, validity and reliability. Moreover, the results provide a deeper view in the way students use the scale to report satisfaction suggesting to revise the instrument according to the suggestion by the National Agency for University Evaluation. Originality/value The paper gives an original contribution on two sides. On the side of methods, it introduces a more accurate methodology for evaluating scales to measure the students’ satisfaction. On the side of applications, it provides important suggestions to the university management to improve the process of quality of the didactics evaluation.

Longitudinal models for dynamic segmentation in financial markets

Purpose Dynamic market segmentation is a very important topic in many businesses where it is interesting to gain knowledge on the reference market and on its evolution over time. Various papers in the reference literature are devoted to the topic and different statistical models are proposed. The purpose of this paper is to compare two statistical approaches to model categorical longitudinal data to perform dynamic market segmentation. Design/methodology/approach The latent class Markov model identifies a latent variable whose states represent market segments at an initial point in time, customers can switch to one segment to another between consecutive measurement occasions and a regression structure models the effects of covariates, describing customers’ characteristics, on segments belonging and transition probabilities. The latent class growth approach models individual trajectories, describing a behaviour over time. Customers’ characteristics may be inserted in the model to affect trajectories that may vary across latent groups, in the author’s case, market segments. Findings The two approaches revealed both suitable for dynamic market segmentation. The advice to marketer analysts is to explore both solutions to dynamically segment the reference market. The best approach will be then judged in terms of fit, substantial results and assumptions on the reference market. Originality/value The proposed statistical models are new in the field of financial markets.

Dynamic segmentation with growth mixture models

This paper proposes a new approach to dynamically segment markets. Dynamic segmentation i of key importance in many markets where it is unrealistic to assume stationary segments due to the dynamics in consumers’ needs and product choices. The main goal of the study is to analyse the dynamic process of financial product ownership under the assumption of heterogeneous growth in different segments taking into account significant determinants of growth trajectories. Using data from 2002 to 2010 collected by the Survey of Household Income and Wealth conducted by the Bank of Italy, this article shows that the Italian market of financial products is segmented and that this behavior’s trajectories over time are significantly influenced by the area of the country where the family lives and head of household’s education and gender.

Identification of latent class Markov models with multiple indicators and correlated measurement errors

A necessary condition for identification of latent class models is that the number of unknown independent parameters must not be greater than the number of observed cells in the contingency table. Such condition is not sufficient at all. Verifying Goodman’s sufficient condition for local identifiability may be, for complex models, a cumbersome procedure. In any case, local identifiability does not guarantee global identifability. The paper provides rules to ascertain global identifiability of some specifications of latent class Markov models, expressing the unknown parameters as a function of the observed frequencies. In the case that not all parameters of a model are identified, the outlined rules provide hints about the restrictions to impose in order to obtain fully identified models.