Document Server@UHasselt
(3.247 recursos)
Repository of the University of Hasselt containing publications in the fields of statistics, computer science, information strategies and material from the Institute for behavioural sciences.
Mostrando recursos 1 - 20 de 36
1.
Type I and type II error under random-effects misspecification in generalized linear mixed models - LITIERE, Saskia; ALONSO ABAD, Ariel; MOLENBERGHS, Geert
Generalized linear mixed models (GLMMs) have become a frequently used tool for the analysis
of non-Gaussian longitudinal data. Estimation is based on maximum likelihood theory, which assumes
that the underlying probability model is correctly specified. Recent research is showing that the results
obtained from these models are not always robust against departures from the assumptions on which these
models are based. In the present work we have used simulations with a logistic random-intercept model
to study the impact of misspecifying the random-effects distribution on the type I and II errors of the
tests for the mean structure in GLMMs. We found that the misspecification can either...
2.
Infant vaccination coverage in 2005 and predictive factors for complete or valid vaccination in Flanders, Belgium: an EPI-survey - Theeten, Heidi; HENS, Niel; Vandermeulen, Corinne; Depoorter, Anne-Marie; Roelants, Mathieu; AERTS, Marc; Hoppenbrouwers, Karel; Van Damme, Pierre
To assess changes in infant vaccination coverage in Flanders since 1999, an EPI-survey was performed in 2005. The parents of 1354 children aged 18-24 months were interviewed at home and the vaccination documents were checked. Several factors possibly related to vaccination status were examined with parametric and non-parametric methods. The coverage rate of recommended vaccines, i.e. poliomyelitis, tetanus-diphtheria-pertussis, H. influenzae type b (Hib), hepatitis B, measles-mumps-rubella (MMR) and meningococcal C, reached at least 92.2%, which is a significant rise for MMR, hepatitis B and Hib since 1999. The vaccinating physician, the employment situation of the mother and the family income...
3.
Handling missingness when modeling the force of infection from clustered seroprevalence data - HENS, Niel; FAES, Christel; AERTS, Marc; SHKEDY, Ziv; MINTIENS, Koen; LAEVENS, Hans; BOELAERT, Frank
Modeling infectious diseases data is a relatively young research area in which clustering and stratification are key features. It is not unlikely for these data to have missing values. If values are missing completely at random, the analysis on the complete cases is valid. However, in practice this assumption is usually not fulfilled. This article shows the effect of ignoring missing data in modeling the force of infection of the bovine herpesvirus-1 in Belgian cattle and proposes the use of weighted generalized estimating equations with constrained fractional polynomials as a flexible modeling tool.
4.
On several two-boundary problems for a particular class of Levy processes - KADANKOVA, Tetyana; VERAVERBEKE, Noel
Several two-boundary problems are solved for a special Levy process: the Poisson process with an exponential component. The jumps of this process are controlled by a homogeneous Poisson process, the positive jump size distribution is arbitrary, while the distribution of the negative jumps is exponential. Closed form expressions are obtained for the integral transforms of the joint distribution of the first exit time from an interval and the value of the overshoot through boundaries at the first exit time. Also the joint distribution of the first entry time into the interval and the value of the process at this time...
5.
Social contacts and mixing patterns relevant to the spread of infectious diseases. - Mossong, Joel; HENS, Niel; Jit, Mark; Beutels, Philippe; Auranen, Kari; et al.
Background :
Mathematical modelling of infectious diseases transmitted by the respiratory or close-contact route (e.g., pandemic influenza) is increasingly being used to determine the impact of possible interventions. Although mixing patterns are known to be crucial determinants for model outcome, researchers often rely on a priori contact assumptions with little or no empirical basis. We conducted a population-based prospective survey of mixing patterns in eight European countries using a common paper-diary methodology.
Methods and Findings :
7,290 participants recorded characteristics of 97,904 contacts with different individuals during one day, including age, sex, location, duration, frequency, and occurrence of physical contact. We found...
6.
A comparison of heterogeneity in the acquisition of varicella zoster virus and parvovirus B19 for five different European countries. - HENS, Niel; GAY, Nigel; SHKEDY, Ziv; AERTS, Marc; EDMUNDS, John
We model the age-dependent force of infection for varicella zoster virus and parvovirus B19 based on a joint model while incorporating individual unobserved heterogeneity. Individual heterogeneity comprises the differences among individuals susceptibility to acquire infections, often referred to as `frailties'. We use a shared gamma frailty to describe this heterogeneity, assuming that the frailty distribution is the same for both infections and contrast this using a correlated frailty approach, relaxing upon this assumption.
7.
Analyzing incomplete discrete longitudinal clinical trial data - JANSEN, Ivy; BEUNCKENS, Caroline; MOLENBERGHS, Geert; VERBEKE, Geert; Mallinckrodt, C
Commonly used methods to analyze incomplete longitudinal clinical trial data include complete case analysis (CC) and last observation carried forward (LOCF). However, such methods rest on strong assumptions, including missing completely at random (MCAR) for CC and unchanging profile after dropout for LOCF. Such assumptions are too strong to generally hold. Over the last decades, a number of full longitudinal data analysis methods have become available, such as the linear mixed model for Gaussian outcomes, that are valid under the much weaker missing at random (MAR) assumption. Such a method is useful, even if the scientific question is in terms...
8.
The nature of sensitivity in monotone missing not at random models - JANSEN, Ivy; HENS, Niel; MOLENBERGHS, Geert; AERTS, Marc; VERBEKE, Geert; Kenward, MG
Models for incomplete longitudinal data under missingness not at random have gained some popularity. At the same time, cautionary remarks have been issued regarding their sensitivity to often unverifiable modeling assumptions. Consequently, there is evidence for a shift towards using ignorable methodology, supplemented with sensitivity analyses to explore the impact of potential deviations of this assumption in the direction of missingness at random. One such tool is local influence. It is shown that local influence tends to pick up a lot of different anomalies in the data at hand, not just deviations in the MNAR mechanism. This particular behavior is...
9.
Analyzing incomplete longitudinal clinical trial data - MOLENBERGHS, Geert; THIJS, Herbert; JANSEN, Ivy; BEUNCKENS, Caroline; Kenward, MG; Mallinckrodt, C; Carroll, RJ
Using standard missing data taxonomy, due to Rubin and co-workers, and simple algebraic derivations, it is argued that some simple but commonly used methods to handle incomplete longitudinal clinical trial data, such as complete case analyses and methods based on last observation carried forward, require restrictive assumptions and stand on a weaker theoretical foundation than likelihood-based methods developed under the missing at random (MAR) framework. Given the availability of flexible software for analyzing longitudinal sequences of unequal length, implementation of likelihood-based MAR analyses is not limited by computational considerations. While such analyses are valid under the comparatively weak assumption of...
10.
Modeling partially incomplete marital satisfaction data - JANSEN, Ivy; Van den Troost, A; MOLENBERGHS, Geert; Vermulst, AA; Gerris, JRM
The authors analyze data on marital satisfaction, obtained from couples at two distinct moments in time (1990, 1995). The data are of a bivariate longitudinal type. Moreover, some couples provide incomplete records only, usually because the 1995 follow-up interview has not taken place. The authors propose a hierarchical modeling strategy that takes all these features into account and is more generally valid than a classical complete case or single imputation-based strategy.
11.
A family of tests to detect misspecifications in the random-effects structure of generalized linear mixed models - ALONSO ABAD, Ariel; LITIERE, Saskia; MOLENBERGHS, Geert
Estimation in generalized linear mixed models for non-Gaussian longitudinal data
is often based on maximum likelihood theory, which assumes that the underlying
probability model is correctly specified. It is known that the results obtained from
these models are not always robust against misspecification of the random-effects
structure. Therefore, diagnostic tools for the detection of this misspecification are
of the utmost importance. Three diagnostic tests, based on the eigenvalues of the
variance-covariance matrices for the fixed-effects parameters estimates, are proposed
in the present work. The power and type I error rate of these tests are studied via
simulations. A very acceptable performance was observed in many cases, especially
for those...
12.
The impact of a misspecified random-effects distribution on the estimation and the performance of inferential procedures in generalized linear mixed models - LITIERE, Saskia; ALONSO ABAD, Ariel; MOLENBERGHS, Geert
Estimation in generalized linear mixed models is often based on maximum likelihood theory, assuming
that the underlying probability model is correctly specified. However, the validity of this assumption is
sometimes difficult to verify. In this paper we study, through simulations, the impact of misspecifying
the random-effects distribution on the estimation and hypothesis testing in generalized linear mixed
models. It is shown that the maximum likelihood estimators are inconsistent in the presence of
misspecification. The bias induced in the mean structure parameters is generally small, as far as
the variability of the underlying random-effects distribution is small as well. However, the estimates
of this variability are always severely...
13.
Type I and type II error under random-effects misspecification in generalized linear mixed models - LITIERE, Saskia; ALONSO ABAD, Ariel; MOLENBERGHS, Geert
Generalized linear mixed models (GLMMs) have become a frequently used tool for the analysis
of non-Gaussian longitudinal data. Estimation is based on maximum likelihood theory, which assumes
that the underlying probability model is correctly specified. Recent research is showing that the results
obtained from these models are not always robust against departures from the assumptions on which these
models are based. In the present work we have used simulations with a logistic random-intercept model
to study the impact of misspecifying the random-effects distribution on the type I and II errors of the
tests for the mean structure in GLMMs. We found that the misspecification can either...
14.
A flexible marginal modelling strategy for non-monotone missing data - JANSEN, Ivy; MOLENBERGHS, Geert
Much research has been devoted to modelling strategies for longitudinal data with missingness, recently especially within the missingness not at random context. In this paper, the relatively unexplored but practically highly relevant domain of non-monotone missingness with multivariate ordinal responses is broached. For this, a dedicated version of the multivariate Dale model is formulated. Furthermore, we also assess the sensitivity of these models to their assumptions, by using the technique of global influence.
15.
Estimating the impact of vaccination using age-time dependent incidence rates of hepatitis B - HENS, Niel; AERTS, Marc; SHKEDY, Ziv; KIMANI, Peter Kung'U; KOJOUHOROVA, Mira; VAN DAMME, Pierre; BEUTELS, Philippe
The objective of this study is to model the age-time dependent incidence of hepatitis B while estimating the impact of vaccination. While stochastic models/time series have been used before to model hepatitis B cases in the absence of knowledge on the number of susceptibles, this paper proposes to use a method that fits into the generalized additive model framework. Generalized additive models with penalized regression splines are used to exploit the underlying continuity of both age and time in a flexible nonparametric way. Based on a unique case notification dataset, it is shown that the implemented immunization programme in Bulgaria...
16.
Reliability of a longitudinal sequence of scale ratings. - LAENEN, Annouschka; ALONSO ABAD, Ariel; MOLENBERGHS, Geert; VANGENEUGDEN, Tony
Longitudinal studies are permeating clinical trials in psychiatry. Addi-
tionally, in the same field, rating scales are frequently used to evaluate the
status of the patients and the efficacy of new therapeutic procedures. There-
fore, it is of utmost importance to study the psychometric properties of these
instruments within a longitudinal framework. In the area of depression, the
Hamilton Depression Rating Scale (HAMD) is regularly used for antidepres-
sant treatment evaluation. However, the use of HAMD has not been exempted
from criticism what has lead to the development of new scales that are ex-
pected to be more sensitive for change, such as the Montgomery-°Asberg De-
pression Rating Scale...
18.
A copula-graphic estimator for the conditional survival function under dependent censoring - BRAEKERS, Roel; VERAVERBEKE, Noel
Rivest & Wells (2001) showed that in situations where the dependence between a lifetime and a censoring variable can be modeled by a given Archimedean copula, the copula-graphic estimator of Zheng & Klein (1995) has an explicit form. The authors extend this work to the fixed design regression case. They show that the copula-graphic estimator then has an asymptotic representation and a Gaussian limit. They also assess the influence of a misspecified copula function on the performance of the estimator. Their developments are illustrated with data on the survival of the Atlantic halibut.
19.
Cox's regression model under partially informative censoring - BRAEKERS, Roel; VERAVERBEKE, Noel
We extend Cox's classical regression model to accomodate partially informative censored data. In this type of data, each observation is the minimum of one lifetime and two censoring times. The survival function of one of these censoring times is a Power of the survival function of the lifetime. We call this the informative censoring time. The distribution of the other censoring time has no relation with the distribution of the lifetime. It is called the non informative censoring time. In this model we specify a semiparametric relation between the lifetime and a covariate where we take into account that also...
20.
Bootstrapping the conditional survival function estimator in the partial Koziol-Green model - BRAEKERS, Roel; VERAVERBEKE, Noel
In the partial Koziol-Green regression model, the lifetime variable may be censored by two types of censoring variables. One is called informative because it satisfies the Koziol-Green assumption on proportionality of hazards and the other one is general. Braekers and Veraverbeke proposed a non-parametric estimator for the conditional lifetime distribution and obtained a Gaussian approximation for the corresponding process. In the present paper, we propose an appropriate resampling scheme and show that this leads to a valid bootstrap approximation for the process.
1.
