Multilevel Analysis of Repeated Measures Data

Hierarchically structured data are common in many areas of scientific research. Such data are characterized by nested membership relations among the units of observation. Multilevel analysis is a class of methods that explicitly takes the hierarchical structure into account. Repeated measures data can be considered as having a hierarchical structure as well: measurements are nested within, for instance, individuals. In this paper, an overview is given of the multilevel analysis approach to repeated measures data. A simple application to growth curves is provided as an illustration. It is argued that multilevel analysis of repeated measures data is a powerful and attractive approach for several reasons, such as flexibility, and the emphasis on individual development.     

An Analysis of WellBeing in Secondary School with Multilevel Growth Curve models and Multilevel Multivariate Models

This paper reports on an exploration of student wellbeing in secondary school. A wellbeing questionnaire was administered four times to the same students. Multilevel models were applied in which measurements are grouped within students within schools. Differences between students are large, but there are only minor differences between schools regarding the wellbeing. Two methods of analysis of longitudinal data are compared: a multilevel multivariate approach and a multilevel growth curve analysis. It is shown that the estimation of individual growth curves is an elegant and parsimonious way of modelling. The multivariate approach on the other hand is a more modest model. The assumptions, advantages and disadvantages of both perspectives are listed.     

M-tests in multivariate models

Summary For testing homogeneity in multivariatek sample model, robust tests based onM-estimators are proposed and their asymptoticx ²-distributions are investigated. FurthermoreM-tests in multivariate regression models are discussed.     

Sensitivity Analysis of Continuous Incomplete Longitudinal Outcomes

Even though models for incomplete longitudinal data are in common use, they are surrounded with problems, largely due to the untestable nature of the assumptions one has to make regarding the missingness mechanism. Two extreme views on how to deal with this problem are (1) to avoid incomplete data altogether and (2) to construct ever more complicated joint models for the measurement and missingness processes. In this paper, it is argued that a more versatile approach is to embed the treatment of incomplete data within a sensitivity analysis. Several such sensitivity analysis routes are presented and applied to a case study, the milk protein trial analyzed before by Diggle and Kenward (1994) . Apart from the use of local influence methods, some emphasis is put on pattern-mixture modeling. In the latter case, it is shown how multiple-imputation ideas can be used to define a practically feasible modeling strategy.     

Regional Inequalities in Consumption Patterns: A Multilevel Approach to the Case of Italy

The main aim of this paper is to evaluate the disparities in the Italian regions on the demand side. In more detail, an attempt will be made to find if the consumption behaviour of Italian households is different in the regions. With this in mind, Istat's 2000 Italian Family Budget data set was analysed. The data in question, which were collected through a two‐stage sample over Italy's 20 regions, contains information regarding the expenses of approximately 23,000 households. In this analysis, both households and regions are considered as units: households are nested in the regions so that the basic data structure is hierarchical. In order to take this hierarchical structure into account, a multilevel model was used, making it possible for parameters to vary randomly from region to region. The model in question also made it possible to consider heterogeneity across different groups (regions), such as stochastic variation. First, regional inequalities were tested using a simple model in which households constituted the first level of analysis and were grouped according to their region (the second level). As a second step, and in order to investigate the interaction between geographical context and income distribution, another model was used. This was cross‐classified by income and regions. The most relevant results showed that there is wide fragmentation of consumption behaviour and, at the same time, various differentiated types of behaviour in the regions under analysis. These territorial differentials become clear from income class and items of consumption.     

Double serial correlation for multilevel growth curve models

Multilevel growth curve models for repeated measures data have become increasingly popular and stand as a flexible tool for investigating longitudinal change in students’ outcome variables. In addition, these models allow the estimation of school effects on students’ outcomes though making strong assumptions about the serial independence of level-1 residuals. This paper introduces a method which takes into account the serial correlation of level-1 residuals and also introduces such serial correlation at level-2 in a complex double serial correlation (DSC) multilevel growth curve model. The results of this study from both real and simulated data show a great improvement in school effects estimates compared to those that have previously been found using multilevel growth curve models without correcting for DSC for both the students’ status and growth criteria.     

On the analysis of multivariate growth curves

Growth curve data arise when repeated measurements are observed on a number of individuals with an ordered dimension for occasions. Such data appear frequently in almost all fields in which statistical models are used, for instance in medicine, agriculture and engineering. In medicine, for example, more than one variable is often measured on each occasion. However, analyses are usually based on exploration of repeated measurements of only one variable. The consequence is that the information contained in the between-variables correlation structure will be discarded.  In this study we propose a multivariate model based on the random coefficient regression model for the analysis of growth curve data. Closed-form expressions for the model parameters are derived under the maximum likelihood (ML) and the restricted maximum likelihood (REML) framework. It is shown that in certain situations estimated variances of growth curve parameters are greater for REML. Also a method is proposed for testing general linear hypotheses. One numerical example is provided to illustrate the methods discussed. Received: 22 February 1999     

Discrete Response Multilevel Models for Repeated Measures: An Application to Voting Intentions Data

Repeated measures data can be modelled as a two-levelmodel where occasions (level one units) are grouped byindividuals (level two units). Goldstein et al. (1994)proposed a multilevel time series model when theresponse variable follows a Normal distribution andthe measurements are taken with unequal timeintervals. This paper extends the methodology todiscrete response variables. The models are applied toBritish Election Study data consisting of repeatedmeasures of voting intention.     

Analysis of longitudinal data using the hierarchical linear model

The hierarchical linear model in a linear model with nested random coefficients, fruitfully used for multilevel research. A tutorial is presented on the use of this model for the analysis of longitudinal data, i.e., repeated data on the same subjects. An important advantage of this approach is that differences across subjects in the numbers and spacings of measurement occasions do not present a problem, and that changing covariates can easily be handled. The tutorial approaches the longitudinal data as measurements on populations of (subject-specific) functions.     

An inflated multivariate integer count hurdle model: an application to bid and ask quote dynamics

In this paper we develop a model for the conditional inflated multivariate density of integer count variables with domain ?ⁿ, n ∈ ?. Our modelling framework is based on a copula approach and can be used for a broad set of applications where the primary characteristics of the data are: (i) discrete domain; (ii) the tendency to cluster at certain outcome values; and (iii) contemporaneous dependence. These kinds of properties can be found for high‐ or ultra‐high‐frequency data describing the trading process on financial markets. We present a straightforward sampling method for such an inflated multivariate density through the application of an independence Metropolis–Hastings sampling algorithm. We demonstrate the power of our approach by modelling the conditional bivariate density of bid and ask quote changes in a high‐frequency setup. We show how to derive the implied conditional discrete density of the bid–ask spread, taking quote clusterings (at multiples of 5 ticks) into account. Copyright © 2009 John Wiley & Sons, Ltd.     

An overlapping generations model with non-ordered preferences and numeraire-incomplete markets*

Abstract This paper shows that, even with a life-cycle component, the standard model of competitive consumption and asset trading can be extended to encompass general preference relations, which do not necessarily hinge upon special assumptions such as time or state separability, or even completeness or transitivity. More precisely, this paper addresses the equilibrium existence for an overlapping generations pure-exchange economy with non-ordered preferences and incomplete financial markets of numeraire assets. Mathematics Subject Classification (2000): 91B50, 91B62 Journal of Economic Literature Classification: D52, D91     

Type I error and test power of different tests for testing interaction effects in factorial experiments

A simulation study was conducted to investigate the effect of non normality and unequal variances on Type I error rates and test power of the classical factorial anova F‐test and different alternatives, namely rank transformation procedure (FR), winsorized mean (FW), modified mean (FM) and permutation test (FP) for testing interaction effects. Simulation results showed that as long as no significant deviation from normality and homogeneity of the variances exists, generally all of the tests displayed similar results. However, if there is significant deviation from the assumptions, the other tests are observed to be affected at considerably high levels except FR and FP tests. As a result, when the assumptions of factorial anova F‐test are not met or, in the case those assumptions are not tested whether met, it can be concluded that using FR and FP tests is more suitable than the classical factorial anova F‐test.     

Comparison of Hypothesis Testing and Bayesian Model Selection

The main goal of both Bayesian model selection and classical hypotheses testing is to make inferences with respect to the state of affairs in a population of interest. The main differences between both approaches are the explicit use of prior information by Bayesians, and the explicit use of null distributions by the classicists. Formalization of prior information in prior distributions is often difficult. In this paper two practical approaches (encompassing priors and training data) to specify prior distributions will be presented. The computation of null distributions is relatively easy. However, as will be illustrated, a straightforward interpretation of the resulting p-values is not always easy. Bayesian model selection can be used to compute posterior probabilities for each of a number of competing models. This provides an alternative for the currently prevalent testing of hypotheses using p-values. Both approaches will be compared and illustrated using case studies. Each case study fits in the framework of the normal linear model, that is, analysis of variance and multiple regression.     

Utilising Recent Advancements in Techniques for the Analysis of Incomplete Multivariate Data to Improve the Data Quality Management of Current Academic Research

This paper discusses the importance of managing data quality in academic research in its relation to satisfying the customer. This focus is on the data completeness objectivedimension of data quality in relation to recent advancements which have been made in the development of methods for analysing incomplete multivariate data. An overview and comparison of the traditional techniques with the recent advancements are provided. Multiple imputation is also discussed as a method of analysing incomplete multivariate data, which can potentially reduce some of the biases which can occur from using some of the traditional techniques. Despite these recent advancements in the analysis of incomplete multivariate data, evidence is presented which shows that researchers are not using these techniques to manage the data quality of their current research across a variety of academic disciplines. An analysis is then provided as to why these techniques have not been adopted along with suggestions to improve the frequency of their use in the future. Source-Reference. The ideas for this paper originated from research work on David J. Fogarty's Ph.D. dissertation. The subject area is the use of advanced techniques for the imputation of incomplete multivariate data on corporate data warehouses.     

Group Status and Entrepreneurship

Do unfettered markets produce too many or too few entrepreneurs? Two seminal papers [ Stiglitz and Weiss (1981) and de Meza and Webb (1987) ] obtained ambiguous answers to this question based on different assumptions about the character of information asymmetries in credit markets. The present paper approaches the same question but using a labor market model in which income is determined by ability and individuals derive utility from income and occupational group status. Occupational group status for entrepreneurs depends on the average entrepreneurial income (due to ex post screening by banks), whereas status for wage employees depends on their own income and ability (due to ex ante screening by employers). Thus, individuals create externalities through their occupational choice. It is shown that there can be too many or too few entrepreneurs in equilibrium depending on the marginal returns to ability in entrepreneurship relative to paid employment; this enables the researcher to use independent evidence about occupational marginal returns to identify the relevant equilibrium likely to arise in practice, together with the likely appropriate policy responses. Based on this approach, we suggest that there may be too many (low ability) entrepreneurs in the USA.     

ECONOMIC WELFARE ANALYSIS OF THE LEGALIZATION OF DRUGS: A CGE MICROSIMULATION MODEL FOR COLOMBIA

In this paper, a computable general equilibrium (CGE) microsimulation model is used to analyze the effects of an ex ante legalization of drugs on the Colombian economy. The model consists of 11 productive sectors, 3 different labor force categories with unemployment, and 20 households divided by income and location. Changes in wages and migration are estimated using a labor participation model, and a NIDS estimates the demands of the households. Changes in household economic welfare, measured by changes in income and prices (CV and EV measurements), are very sensitive to the reinvestments that the government makes in the economy. By analyzing six different scenarios with different assumptions about changes in drug prices, investments of the government, and the termination of the armed conflict, the results suggest that economic welfare improves when the government reinvests military expenditures in other productive sectors or when the ‘economy of war’ continues and the legalization does not end the armed conflict.     

Is gravity linear?

Despite the solid theoretical foundation on which the gravity model of bilateral trade is based, empirical implementation requires several assumptions which do not follow directly from the underlying theory. First, unobserved trade costs are assumed to be a (log‐)linear function of observables. Second, the effects of trade costs on trade flows are assumed to be constant across country pairs. Maintaining consistency with the underlying theory, but relaxing these assumptions, we estimate gravity models—in levels and logs—using two data sets via nonparametric methods. The results are striking. Despite the added flexibility of the nonparametric models, parametric models based on these assumptions offer equally or more reliable in‐sample predictions and out‐of‐sample forecasts in the majority of cases, particularly in the levels model. Moreover, formal statistical tests fail to reject either parametric functional form. Thus, concerns in the gravity literature over functional form appear unwarranted, and estimation of the gravity model in levels is recommended. Copyright © 2008 John Wiley & Sons, Ltd.     

Efficiency vectors, efficiency and performance measures: new methods for ranking efficient organizational units

We present and discuss measures for analyzing productivity in deterministic frontier models. A new efficiency measure is introduced allowing for discrimination among efficient organizational units. In addition, a new performance measure for analyzing productivity of organizational units is presented. This measure fulfills various properties of efficiency measures but relaxes the indicator property. Both new approaches are based on the development of efficiency vectors which is a new vector measure for measuring efficiency. The vector components are efficiency measures related to subsets of a production possibility set. The new approaches are applied in the context of data envelopment analysis.

Influence analysis for Poisson inverse Gaussian regression models based on the EM algorithm

For Poisson inverse Gaussian regression models, it is very complicated to obtain the influence measures based on the traditional method, because the associated likelihood function involves intractable expressions, such as the modified Bessel function. In this paper, the EM algorithm is employed as a basis to derive diagnostic measures for the models by treating them as a mixed Poisson regression with the weights from the inverse Gaussian distributions. Several diagnostic measures are obtained in both case-deletion model and local influence analysis, based on the conditional expectation of the complete-data log-likelihood function in the EM algorithm. Two numerical examples are given to illustrate the results.