BCC

Multivariate and mixed linear models

13.06.2021 - 19.06.2021 | Będlewo & Online (hybrid)

Programme

On-line session:

Wednesday, 16.06.2021:

16:00 Opening

16:05 Dietrich von Rosen "Classification with more than two populations"

16:40 Malwina Janiszewska "The condition for the sample size"

break

17:00 Dário Ferreira "Inference in prime factorial models"

17:30 Ivan Žežula "More on mean testing in three-level data"

 

Thursday, 17.06.2021

 

10:00 Jolanta Pielaszkiewicz "Relative importance in logistic regression"

10:30 Adam Mieldzioc "The comparison of the covariance matrix estimators under banded Toeplitz structure"

break

11:00 Daniel Klein "Testing the compond symmetry structure in large- and high-dimensional setting"

11:30 Katarzyna Filipiak "Estimation of parameters under the growth curve model with block random effects"

 

 

More details:

Dário Ferreira "Inference in prime factorial models"

Authors: Dário Ferreira, Sandra Ferreira, Célia Nunes and João T. Mexia

AbstractThe use of factorial designs are preferable in experiments that involve the study of the effects of two or more factors since they are distinguished for their flexibility. These models are widely used, for exemple in fertilization trials. In this work we show how to simplify and carry out such experiments. We illustrate the theory with an application to real data.
References:
[1] R.A. Bailey, S.S. Ferreira, D. Ferreira, C. Nunes, Estimability of Variance Components when all Model Matrices Commute, Linear Algebra and its Applications, 492, 144 160, 2015.
[2] F. Carvalho, Strictly associated models, prime basis factorials: an application, Discussiones Mathematicae Probability and Statistics, 31, 77 86, 2011.
[3] D. Ferreira, S.S. Ferreira, C. Nunes, J.T. Mexia, Tests and relevancies for the hypotheses of an orthogonal family in a model with orthogonal block structure, Journal of Statistical Computation and Simulation, 90, 3, 412{419, 2020.
[4] S. Oliveira, C. Nunes, E. Moreira, M. Fonseca and J. T. Mexia, Balanced prime basis factorial fixed effects model with random number of observations, Journal of Applied Statistics, 2019, Doi: https://doi.org/10.1080/02664763.2019.1679097
Acknowledgements: This work was partially supported by the Portuguese Foundation for Science and Technology through the projects UIDP/MAT/00212/2020 and UIDP/MAT/00297/2020.

 

 

 

Rewrite code from the image

Reload image

Reload image