
As an Amazon Associate and affiliate partner, Menrva Books earns from qualifying purchases. Learn more
Spatial statistics plays an important role in various fields, where data are observed with location information, including environmental science, geography and epidemiology. Such data are usually modeled through spatial random process. In this research, we investigate spatial random process in both Euclidean space and on the circle.In Euclidean space, variogram is popularly used to study the spatial dependence. A valid variogram needs to be conditionally negative definite. In this research, we obtain novel inversion formulae between variogram and its associated spectral function. Then, a valid nonparametric variogram estimator is proposed based on these inversions through the constrained optimization. We demonstrate this approach through simulation studies.On the circle, a proper probability space for the white noise is investigated. We first demonstrate that the commonly applied Brownian motion on the circle constructed by Levy is problematic. We show that it is a regular Euclidean Brownian motion on the half-circle with its own mirror image on the half-circle, and is degenerated in the sense of Minlos's study. We then formally define the white noise space and its associated Brownian bridge on the circle.
This research investigates spatial random processes, focusing on nonparametric variogram estimation in Euclidean space and the definition of white noise space on a circle. The author, Ao Li, a statistician, addresses the importance of spatial statistics in fields like environmental science and geography. In Euclidean space, the work derives novel inversion formulae between variograms and their spectral functions, leading to a constrained optimization-based nonparametric variogram estimator, validated through simulations. On the circle, the research scrutinizes the construction of Brownian motion by Lévy, identifying it as a regular Euclidean Brownian motion on a half-circle with a mirrored image, and consequently degenerate. This leads to a formal definition of the white noise space and its associated Brownian bridge on the circle.
The research presented in this work offers novel theoretical contributions to spatial statistics. The development of nonparametric variogram estimators through constrained optimization in Euclidean space, supported by simulation studies, indicates a practical approach to analyzing spatial dependence. Furthermore, the critical examination and formal definition of white noise space and Brownian bridges on the circle address foundational aspects of stochastic processes in non-Euclidean settings. This work is likely to be of interest to researchers and advanced students in statistics, mathematics, and related applied fields.
Page Count:
58
Publication Date:
2020-01-01
Publisher:
Indiana University
ISBN-13:
9798672192062
No comments yet. Be the first to share your thoughts!