Premium
Discussion on A high‐resolution bilevel skew‐ t stochastic generator for assessing Saudi Arabia's wind energy resources
Author(s) -
ZammitMangion Andrew
Publication year - 2020
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/env.2649
Subject(s) - computer science , inference , skew , statistical inference , domain (mathematical analysis) , data science , big data , wind power , generator (circuit theory) , focus (optics) , construct (python library) , industrial engineering , operations research , data mining , artificial intelligence , mathematics , power (physics) , statistics , engineering , telecommunications , mathematical analysis , physics , optics , quantum mechanics , electrical engineering , programming language
Statistical spatiotemporal environmental data analysis is rarely straightforward, with one having to face challenges relating to big data, non‐Gaussianity, nonstationarity, multiple scales of behavior, deterministic (numerical) model output, and more. One often has to rely heavily on good statistical parallel computing skills and sound knowledge of the application domain. The work of Tagle et al. (2020) overcomes all of these challenges, and is an excellent example of the tangible contributions spatiotemporal modeling and distribution theory can make to the environmental sciences at the policy level. In this discussion piece I focus on a few high‐level concepts in the paper of Tagle et al. (2020) that are relevant to related application domains. I also provide some technical suggestions that could be used to facilitate inference.