figshare
Browse
DOCUMENT
uasa_a_1775613_sm9761.pdf (216.56 kB)
TEXT
uasa_a_1775613_sm9760.r (11.89 kB)
1/0
2 files

Detection of Local Differences in Spatial Characteristics Between Two Spatiotemporal Random Fields

Version 2 2020-08-24, 09:30
Version 1 2020-06-03, 15:54
dataset
posted on 2020-08-24, 09:30 authored by Sooin Yun, Xianyang Zhang, Bo Li

Comparing the spatial characteristics of spatiotemporal random fields is often at demand. However, the comparison can be challenging due to the high-dimensional feature and dependency in the data. We develop a new multiple testing approach to detect local differences in the spatial characteristics of two spatiotemporal random fields by taking the spatial information into account. Our method adopts a two-component mixture model for location wise p-values and then derives a new false discovery rate (FDR) control, called mirror procedure, to determine the optimal rejection region. This procedure is robust to model misspecification and allows for weak dependency among hypotheses. To integrate the spatial heterogeneity, we model the mixture probability as well as study the benefit if any of allowing the alternative distribution to be spatially varying. An EM-algorithm is developed to estimate the mixture model and implement the FDR procedure. We study the FDR control and the power of our new approach both theoretically and numerically, and apply the approach to compare the mean and teleconnection pattern between two synthetic climate fields. Supplementary materials for this article are available online.

Funding

Yun and Li’s research was partially supported by National Science Foundation grant AGS-1602845 and DMS-1830312. Zhang acknowledges partial support from NSF DMS-1830392 and NSF DMS-1811747.

History