liu.seSearch for publications in DiVA
Change search
Link to record
Permanent link

Direct link
Byukusenge, Béatrice
Publications (3 of 3) Show all publications
Byukusenge, B., von Rosen, D. & Singull, M. (2023). On Residual Analysis in the GMANOVA-MANOVA Model. In: Balakrishnan, Narayanaswamy; Gil, María Ángeles; Martín, Nirian; Morales, Domingo; Pardo, María del Carmen (Ed.), Trends in Mathematical, Information and Data Sciences: A Tribute to Leandro Pardo: (pp. 287-305). Springer International Publishing
Open this publication in new window or tab >>On Residual Analysis in the GMANOVA-MANOVA Model
2023 (English)In: Trends in Mathematical, Information and Data Sciences: A Tribute to Leandro Pardo / [ed] Balakrishnan, Narayanaswamy; Gil, María Ángeles; Martín, Nirian; Morales, Domingo; Pardo, María del Carmen, Springer International Publishing , 2023, p. 287-305Chapter in book (Refereed)
Abstract [en]

In this article, the GMANOVA-MANOVA model is considered. Two different matrix residuals are established. The interpretation of the residuals is discussed and several properties are verified. A data set illustrates how the residuals can be used.

Place, publisher, year, edition, pages
Springer International Publishing, 2023
Series
Studies in Systems, Decision and Control, E-ISSN 2198-4190 ; 445
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:liu:diva-189840 (URN)10.1007/978-3-031-04137-2_24 (DOI)9783031041365 (ISBN)9783031041372 (ISBN)
Available from: 2022-11-09 Created: 2022-11-09 Last updated: 2023-01-26Bibliographically approved
Byukusenge, B., von Rosen, D. & Singull, M. (2022). On the Identification of Extreme Elements in a Residual for the GMANOVA-MANOVA Model. In: Andriëtte Bekker, Johannes T. Ferreira, Mohammad Arashi and Ding-Geng Chen (Ed.), Innovations in Multivariate Statistical Modeling: Navigating Theoretical and Multidisciplinary Domains (pp. 119-135). Cham: Springer Cham
Open this publication in new window or tab >>On the Identification of Extreme Elements in a Residual for the GMANOVA-MANOVA Model
2022 (English)In: Innovations in Multivariate Statistical Modeling: Navigating Theoretical and Multidisciplinary Domains / [ed] Andriëtte Bekker, Johannes T. Ferreira, Mohammad Arashi and Ding-Geng Chen, Cham: Springer Cham , 2022, p. 119-135Chapter in book (Refereed)
Abstract [en]

Two different matrix residuals in a special GMANOVA-MANOVA model have previously been established (see Byukusenge et al., 2021, “On residual analysis in the GMANOVA-MANOVA model”). The residual that is studied in this article is constructed via the difference of the observed group means and the estimated mean structure. The residual provides information about the appropriateness of the model assumptions concerning the mean structure. The aim of this paper is to study the distribution of the largest elements (by absolute value) of the residual via two data sets. Parametric bootstrap is used to identify thresholds so that extreme elements of the residuals can be identified.

Place, publisher, year, edition, pages
Cham: Springer Cham, 2022
Series
Emerging Topics in Statistics and Biostatistics, ISSN 2524-7735, E-ISSN 2524-7743 ; 1
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:liu:diva-198562 (URN)10.1007/978-3-031-13971-0_6 (DOI)9783031139710 (ISBN)
Available from: 2023-10-17 Created: 2023-10-17 Last updated: 2023-11-21Bibliographically approved
Byukusenge, B. (2022). Residual Analysis in the GMANOVA-MANOVA Model. (Doctoral dissertation). Linköping: Linköping University Electronic Press
Open this publication in new window or tab >>Residual Analysis in the GMANOVA-MANOVA Model
2022 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This thesis focuses on the establishment and analysis of residuals in the so called GMANOVA-MANOVA model. The model is a special case of the Extended Growth Curve Model. It has two terms where one term models the profiles (growth curves) and the other the covariables of interest. This model is useful in studying growth curves in short time series in fields such as economics, biology, medicine, and epidemiology. Furthermore, in the literature, residuals have been extensively studied and used to check model adequacy in univariate linear models. This thesis contributes to the extension of the study of residuals in the GMANOVA-MANOVA model. 

In this thesis, a new pair of residuals is established via the maximum likelihood estimators of the parameters in the model. One residual indicates whether an individual is far away from the group means and a second residual is used to check assumptions about the mean structure. Different properties of these residuals are verified and their interpretation is discussed. Moreover, using parametric bootstrap, the empirical distributions of the extreme elements in the residuals are derived. 

Finally, testing bilinear restriction in the MANOVA model is considered. One can show that the MANOVA model with bilinear restrictions is nothing more than a GMANOVA-MANOVA model. Furthermore, the likelihood ratio test can be shown to be given as a function of the residuals to the GMANOVA-MANOVA model, which can be used to understand the appropriateness of the model and test the bilinear hypothesis. 

Abstract [sv]

I den här avhandling härleds och analyseras residualer för den så kallade GMANOVA-MANOVA modellen. Modellen, som är ett specialfall av den utökade tillväxtkurvemodellen, har två delar där den ena beskriver profilerna (väntevärdesmodellen som tillväxtkurvor) för olika grupper och den andra delen i modellen de kovariater som kan vara av vikt (variabler som påverkar analysen men som inte direkt är av intresse). GMANOVA-MANOVA modellen är användbar för att studera upprepade mätningar, som korta tidsserier, inom områden som teknik, ekonomi, biologi, medicin och epidemiologi.

Residualer för en statistisk modell är ett viktigt verktyg för att studera om modellantaganden är uppfyllda. Den här avhandlingen bidrar till resiudalanalys i GMANOVA-MANOVA-modellen, genom att härleda två residualer för modellen. Den första residualen beskriver om en observation är långt borta från gruppmedelvärdet medan den andra residualen används för att kontrollera antaganden om väntevärdesmodellen (profilen) är uppfyllda. Olika egenskaper hos dessa residualer härleds och deras tolkning diskuteras. Dessutom, beskrivs med hjälp av parametrisk bootstrap, hur de empiriska fördelningarna av de extrema elementen i residualerna kan beräknas och utnyttjas i analysen.  

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2022. p. 43
Series
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 2221
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:liu:diva-189841 (URN)10.3384/9789179292850 (DOI)9789179292843 (ISBN)9789179292850 (ISBN)
Public defence
2022-12-09, BL32 (Nobel) B-building, Campus Valla, Linköping, 10:15 (English)
Opponent
Supervisors
Available from: 2022-11-09 Created: 2022-11-09 Last updated: 2022-11-09Bibliographically approved
Organisations

Search in DiVA

Show all publications