On uncertainty estimation in functional linear mixed models. Issue 3 (26th November 2020)
- Record Type:
- Journal Article
- Title:
- On uncertainty estimation in functional linear mixed models. Issue 3 (26th November 2020)
- Main Title:
- On uncertainty estimation in functional linear mixed models
- Authors:
- Maiti, Tapabrata
Safikhani, Abolfazl
Zhong, Ping‐Shou - Abstract:
- Abstract: Functional data analysis has proven useful in many scientific applications where a physical process is observed as a curve. In many applications, several curves are observed due to multiple subjects, providing replicates in the statistical sense. Recent literature develops several techniques for registering curves and estimating associated models in a regression framework. Standard regression models ignore heterogeneity among curves. Functional linear mixed models are one popular way to combine several curves and capture variability among curves via random effects. Although there is a good amount of work available for analyzing functional data using mixed models, limited attention has been paid to inference. After estimation, we concentrate on measuring uncertainty in terms of mean squared error when functional linear mixed models are used for prediction. Although measuring uncertainty is of paramount interest in any statistical prediction, there is no theoretically valid expression available for functional mixed effect models. The quality of theoretical approximations depends on the number of curves observed. In many real life applications, only a finite number of curves can be observed. In such situations, it is important to asses the error rate for any valid statistical statement. In this article, we derive a theoretically valid approximation of uncertainty measurements for prediction. We also provide some modifications for model estimation. The empiricalAbstract: Functional data analysis has proven useful in many scientific applications where a physical process is observed as a curve. In many applications, several curves are observed due to multiple subjects, providing replicates in the statistical sense. Recent literature develops several techniques for registering curves and estimating associated models in a regression framework. Standard regression models ignore heterogeneity among curves. Functional linear mixed models are one popular way to combine several curves and capture variability among curves via random effects. Although there is a good amount of work available for analyzing functional data using mixed models, limited attention has been paid to inference. After estimation, we concentrate on measuring uncertainty in terms of mean squared error when functional linear mixed models are used for prediction. Although measuring uncertainty is of paramount interest in any statistical prediction, there is no theoretically valid expression available for functional mixed effect models. The quality of theoretical approximations depends on the number of curves observed. In many real life applications, only a finite number of curves can be observed. In such situations, it is important to asses the error rate for any valid statistical statement. In this article, we derive a theoretically valid approximation of uncertainty measurements for prediction. We also provide some modifications for model estimation. The empirical performance of the proposed method is investigated by numerical examples and is compared with existing literature as appropriate. Our method is computationally simple and often outperforms other existing methods. Résumé : L'analyse fonctionnelle s'avère utile pour de nombreuses applications scientifiques où un procédé physique est observé sous la forme d'une courbe. Pour bon nombre d'applications, plusieurs courbes sont observées car de multiples sujets sont mesurés, ce qui produit des répétitions au sens statistique. Dans la littérature récente, plusieurs techniques sont développées pour l'enregistrement des courbes et l'estimation des modèles associés dans le cadre d'une régression. Les modèles habituels ignorent l'hétérogénéité parmi les courbes. Les modèles linéaires fonctionnels mixtes offrent une façon populaire de combiner de nombreuses courbes et de capturer la variabilité entre elles par des effets aléatoires. Malgré la grande quantité de travail consacré à l'analyse de données fonctionnelles avec des modèles mixtes, les questions d'inférence ont reçu peu d'attention. Après l'estimation, les auteurs s'affairent à mesurer l'incertitude en termes d'erreur quadratique moyenne lorsque les modèles linéaires fonctionnels mixtes servent à la prédiction. Même si la mesure de l'incertitude est d'un intérêt primordial pour toute prévision statistique, aucune expression théorique valide n'est disponible pour les modèles fonctionnels mixtes. La qualité des approximations théoriques dépend du nombre de courbes observées, et pour plusieurs applications réelles, seul un nombre fini de courbes peut être observé. Dans de telles circonstances, il est important d'évaluer le taux d'erreur pour chaque énoncé statistique valide. Les auteurs dérivent une approximation théoriquement valide des mesures d'incertitude pour la prévision. Ils apportent également quelques modifications à l'estimation de modèle. Ils investiguent les performances empiriques de la méthode proposée par des exemples numériques et la comparent aux méthodes existantes lorsque c'est approprié. Ils constatent que leur méthode est simple d'un point de vue computationnel et qu'elle offre souvent de meilleures performances que les méthodes existantes. … (more)
- Is Part Of:
- Canadian journal of statistics. Volume 49:Issue 3(2021)
- Journal:
- Canadian journal of statistics
- Issue:
- Volume 49:Issue 3(2021)
- Issue Display:
- Volume 49, Issue 3 (2021)
- Year:
- 2021
- Volume:
- 49
- Issue:
- 3
- Issue Sort Value:
- 2021-0049-0003-0000
- Page Start:
- 771
- Page End:
- 792
- Publication Date:
- 2020-11-26
- Subjects:
- Basis functions -- B‐splines -- bias correction -- estimating equations -- functional mixed models -- Karhunen–Loève expansion
Mathematical statistics -- Periodicals
519.5 - Journal URLs:
- http://archimede.mat.ulaval.ca/cjs/ ↗
http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1708-945X/issues ↗
http://www.jstor.org/journals/03195724.html ↗
http://onlinelibrary.wiley.com/ ↗
http://www.ingentaconnect.com/content/ssc/cjs ↗
http://www.mat.ulaval.ca/rcs/indexe.shtml ↗ - DOI:
- 10.1002/cjs.11585 ↗
- Languages:
- English
- ISSNs:
- 0319-5724
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 3035.760000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 24519.xml