Technical advances in agricultural science build on large and often complex information sources. This information can be optimally exploited with appropriate statistical models and computational tools. This PhD project addresses statistical modeling of longitudinal data - the data recorded on individuals are collected repeatedly over time (or space) - from agricultural applications.

Smitha Ankinakatte in her PhD study has focused on modeling two kinds of longitudinal data: firstly, multivariate continuous time series data obtained from robotic milking of cows and secondly, univariate discrete data arises for example in animal and plant genetics.

Part I of the work aims to develop and compare two statistical methods to predict mastitis, neural networks and generalized additive models. These use multivariate indicators of mastitis that are recorded online in an automatic milking system.

Part II of the study concerns the statistical modeling of longitudinal univariate discrete data, with particular focus on the data generated by SNP arrays. An R package, gRapfa, has been developed to implement the methods.

Time: Tuesday 13 May 2014 at 12.00
Place: The Auditorium, AU Foulum
Title of dissertation: Statistical Methods for longitudinal data with agricultural applications
Contact information: Smitha Ankinakatte, e-mail: Smitha.AA@agrsci.dk, tel.: +45 8716 9281
Members of the assessment committee:
Associate professor. Poul Svante Eriksen, Department of Mathematical Sciences, Aalborg University
Dr. Ir. Scientific researcher Rudi de Mol, Animal Science Group, Wageningen UR
Professor Just Jensen (chair), Department of Molecular Biology and Genetics
Main supervisor:
David Edwards, Center for Quantitative Genetics and Genomics, Dept. of Molecular Biology and Genetics, Aarhus University
Co-supervisor:
Ole Fredslund Christensen, Center for Quantitative Genetics and Genomics, Dept. of Molecular Biology and Genetics, Aarhus University
Language: The PhD dissertation will be defended in English.

The defence is public.
The dissertation is available for reading at the Graduate School of Science and Technology/GSST, Ny Munkegade 120, building 1521, room 112, 8000 Aarhus C.