An Overview of Variance Component Estimation.
We consider a chaotic system with a double-scroll attractor proposed by Elwakil, composing with a second-order system, which has low-dimensional multiple invariant subspaces and multi-level on-off intermittency. This type of composite system always includes a skew-product structure and some invariant subspaces, which are associated with different levels of laminar phase. In order for the level of laminar phase be adjustable, we adopt a nonlinear function with saturation characteristic to tune the...
Let us consider the linear model covering the one-way classification as a special casse. In the paper the relationship between testing of some linear hypothesis and estimating of parameters in the linear model by common software packages is examined.
Planteamos el modelo lineal sobre variables dummy difusas asociadas a variables cuantitativas codificadas o cualitativas. En el modelo bicriterio aparecen las mismas dificultades conceptuales que se presentan en el caso multicriterio, pero su desarrollo resulta mucho más ágil y transparente. Resolvemos las ecuaciones normales, caracterizando el conjunto de soluciones y encontramos finalmente las funciones paramétricas estimables para este modelo.
Continuando el desarrollo del modelo lineal sobre dos variables dummy difusas, encontramos expresiones para las descomposiciones clásicas en sumas de cuadrados y estudiamos sus distribuciones bajo hipótesis de normalidad. Damos una interpretación geométrica del ajuste cuando las variables numéricas explicativas se ven sometidas a una codificación difusa de tipo semilineal y contrastamos en estos casos la mejora de este nuevo modelo respecto de la regresión sobre las variables originales. Concluimos...
Binary operations on commutative Jordan algebras are used to carry out the ANOVA of a two layer model. The treatments in the first layer nests those in the second layer, that being a sub-model for each treatment in the first layer. We present an application with data retried from agricultural experiments.
For estimating the variance components of a one-way random effect model recently Uhlig (1995, 1997) and Lischer (1996) proposed non-iterative estimators with high breakdown points. These estimators base on the high breakdown point scale estimators of Rousseeuw and Croux (1992, 1993), which they called Q-estimators. In this paper the asymptotic normal distribution of the new variance components estimators is derived so that the asymptotic efficiency of these estimators can be compared with that of...