Comments on: Data science, big data and statistics
- 27 Downloads
We would like to start by congratulating the authors for a very timely and stimulating paper. They have provided thought-provoking ideas on Data Science and Big Data, and on how Statistics must play a major role in these new areas. We focus our discussion on two points that have caught our attention and interest: visualization and computations for new sources of information.
Visualization for new sources of information
Traditionally, Statistics has dealt with scalar and vectorial observations. However, as noted by the authors, advances in technology have greatly facilitated the collection of large-scale high-dimensional data in many research fields. Among various types of high-dimensional data, spatiotemporal data and functional data have been particularly popular. Classical statistical methodologies face many challenges for such datasets because they often contain massive amounts of observations, non-Gaussian features, and they may exhibit complex spatiotemporal dynamics....
Mathematics Subject Classification62M30 62H30
- Abdulah S, Ltaief H, Sun Y, Genton MG, Keyes DE (2018a) Parallel approximation of the maximum likelihood estimation for the prediction of large-scale geostatistics simulations. In: IEEE Int Conf Clust Comput, pp 98–108Google Scholar
- Euán C, Sun Y (2019) Directional spectra-based clustering methods for visualizing patterns of winds and waves in the Red Sea. J Comput Graph Stat. https://doi.org/10.1080/10618600.2019.1575745
- Euán C, Ombao H, Ortega J (2018) The hierarchical spectral merger algorithm: a new time series clustering procedure. J Classif 35:71–99Google Scholar
- Euán C, Sun Y, Ombao H (2019) Coherence-based time series clustering for statistical inference and visualization of brain connectivity. Ann Appl Stat (to appear)Google Scholar
- Huang H, Sun Y (2019) Visualization and assessment of spatio-temporal covariance properties. Spat Stat. https://doi.org/10.1016/j.spasta.2017.11.004
- Sun Y, Li B, Genton MG (2012b) Geostatistics for large datasets, Chap 3. In: Porcu E, Montero JM, Schlather M (eds) Space-time processes and challenges related to environmental problems, vol 207. Springer, Berlin, pp 55–77Google Scholar