Linear mixed model with Laplace distribution (LLMM)

2018-03-01
Linear mixed modeling (LMM) is a comprehensive technique used for clustered, panel and longitudinal data. The main assumption of classical LMM is having normally distributed random effects and error terms. However, there are several situations for that we need to use heavier tails distributions than the (multivariate) normal to handle outliers and/or heavy tailness in data. In this study, we focus on LMM using the multivariate Laplace distribution which is known as the heavy tailed alternative to the normal distribution. The parameter estimators of interest are generated with EM algorithm for the proposed model. A simulation study is provided to illustrate the performance of the Laplace distribution over the normal distribution for LMM. Also, a real data example is used to explore the behavior of the proposed estimators over the counterparts.
STATISTICAL PAPERS

Suggestions

Failure analysis of tapered composite structures under tensile loading
Çelik, Ozan; Parnas, Kemal Levend; Department of Mechanical Engineering (2016)
A three dimensional finite element modeling approach is used to evaluate the effects of preliminary design variables on the performance of tapered composite laminates under tensile loading. Hashin failure criteria combined with a progressive failure algorithm is used for in-plane failure mechanisms and cohesive zone method is used for out-of-plane failures. The modeling approach is validated by a comparison with experimental results from literature. The validated model is used to examine various design vari...
Sliding mode control design with time varying sliding surfaces for a class of nonlinear systems
SALAMCİ, METİN UYMAZ; Tombul, G. Serdar (2006-10-06)
This paper presents sliding mode control (SMC) design for a class of nonlinear systems in which the sliding surface is designed to be linear with time varying slope(s). The sliding surface design is based on the frozen-time approach. The nonlinear system is frozen at each operation step resulting in linear time invariant (LTI) model and the sliding surface is designed for the LTI model. The surface slope is updated at each frozen step which gives, in general, a moving sliding surface. The control term, on t...
Real-Time Detection of Interharmonics and Harmonics of AC Electric Arc Furnaces on GPU Framework
Uz-Logoglu, Eda; Salor, Ozgul; Ermiş, Muammer (2017-10-05)
In this paper, a method based on the multiple synchronous reference frame (MSRF) analysis is recommended and implemented to detect time-varying harmonics and interharmonics of rapidly fluctuating asymmetrical industrial loads. The experimental work has been carried out on a typical three-phase alternating current arc furnace (AC EAF) installation. In the recommended method, the reference frame is rotated in both directions at speeds corresponding to the positive and negative sequences of all harmonics and a...
Real-Time Detection of Interharmonics and Harmonics of AC Electric Arc Furnaces on GPU Framework
Uz-Logoglu, Eda; Salor, Ozgul; Ermiş, Muammer (2019-11-01)
In this paper, a method based on the multiple synchronous reference frame analysis is recommended and implemented to detect time-varying harmonics and interharmonics of rapidly fluctuating asymmetrical industrial loads. The experimental work has been carried out on a typical three-phase alternating current arc furnace installation. In the recommended method, the reference frame is rotated in both directions at speeds corresponding to the positive and negative sequences of all harmonics and all interharmonic...
Lateral stiffness estimation in frames and its implementation to continuum models for linear and nonlinear static analysis
EROĞLU AZAK, TUBA; Akkar, Dede Sinan (2011-08-01)
Continuum model is a useful tool for approximate analysis of tall structures including moment-resisting frames and shear wall-frame systems. In continuum model, discrete buildings are simplified such that their overall behavior is described through the contributions of flexural and shear stiffnesses at the story levels. Therefore, accurate determination of these lateral stiffness components constitutes one of the major issues in establishing reliable continuum models even if the proposed solution is an appr...
Citation Formats
F. Gökalp Yavuz, “Linear mixed model with Laplace distribution (LLMM),” STATISTICAL PAPERS, pp. 271–289, 2018, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/35371.