Advanced Search

Consider an infinite dimensional diffusion process process on $T^{{𝐙}^d}$, where $T$ is the circle, defined by the action of its generator $L$ on $C^2\left(T^{{𝐙}^d}\right)$ local functions as $Lf\left(\eta \right)={\sum }_{i\in {𝐙}^d}\left(\frac{1}{2}{a}_i\frac{{\partial }^{2}f}{\partial {\eta }_i^2}+b_i\frac{\partial f}{\partial {\eta }_i}\right)$. Assume that the coefficients, $a_i$ and $b_i$ are smooth, bounded, finite range with uniformly bounded second order partial derivatives, that $a_i$ is only a function of ${\eta }_i$ and that ${inf}_{i,\eta }{a}_i\left(\eta \right)\>0$. Suppose $\nu$ is an invariant product measure. Then, if $\nu$ is the Lebesgue measure or if $d=1,2$, it is the unique invariant measure. Furthermore, if $\nu$ is translation invariant, then...

Consider an infinite dimensional diffusion process process on , where is the circle, defined by the action of its generator on ) local functions as $Lf\left(\eta \right)={\sum }_{i\in {𝐙}^d}\left(\frac{1}{2}{a}_i\frac{{\partial }^{2}f}{\partial {\eta }_i^2}+b_i\frac{\partial f}{\partial {\eta }_i}\right)$. Assume that the coefficients, and are smooth, bounded, finite range with uniformly bounded second order partial derivatives, that is only a function of ${\eta }_i$ and that ${inf}_{i,\eta }{a}_i\left(\eta \right)>0$. Suppose is an invariant product measure. Then, if is the Lebesgue measure or if , it is the unique...

Para varios conjuntos de datos biomédicos, se quiere hacer inferencia sobre la función media de la población. La mayoría de los métodos clásicos para este tipo de inferencia han sido derivados bajo el supuesto de que los datos constituyen muestras aleatorias de una población normal. Pero en realidad, en el campo médico, pocas veces se tienen muestras aleatorias. El problema práctico principal es hallar un modelo de dependencia entre los datos que aproxime la situación real y que sea manejable desde...