J. M. Rodríguez Díaz, G. Sánchez León
When measuring different types of responses from an experiment, a covariance structure among samples taken at the same point x is usually assumed, while the observations taken at different points are always assumed independent. The last assumption, that could appear as reasonable when the variable X refers to individuals in a study, seems too risky in some other cases, for instance when dealing with samples taken on the same individual at distinct temporal points. Thus, apart from the 'static' covariance structure among samples of different kind taken at the same point (intra-covariance), we should consider as well the 'longitudinal' covariance structure among same-type samples taken at different points (inter-covariance). This fact turns the optimal design study into a cumbersome problem, with a high computational cost. In order to ease computations some analytical results are shown, as well as different examples of application.
Palabras clave / Keywords: correlation, covariance matrix, linear models, optimal design of experiments
Programado
Sesión GT07-1 Diseño de Experimentos: Advances in Experimental Design (OED-1). Organizadores: Víctor Casero-Alonso y Jesús López-Fidalgo
30 de mayo de 2018 10:50
Sala 6
