Rejoinder on: Data science, big data and statistics
- 14 Downloads
We are very grateful to all the discussants for agreeing to comment on our paper, for their encouraging words about our work and for their wise insights and ideas provided in their contributions. Their comments have broadened the scope of our presentation and have enriched the relationships among data science, big data and statistics with many useful references.
We fully agree on the importance of image data in health applications emphasized by Prof. Bühlmann, and we expect more use of image data in all fields of science. Prof. Bühlmann also stressed an important point that we do not cover in our article: stability analysis of the model, or prediction rule, fitted to the data. With this objective, we can estimate the model in different subsamples, that in particular could be different clusters, or create new data sets by perturbation of the original sample with some distribution. This is an important area of research and is connected to the comments of Prof. Delicado, on model...
Mathematics Subject Classification62A01 62H99
- Alonso AM, Galeano P, Peña D (2019) A robust procedure to build dynamic factor models with cluster structure (manuscript submitted) Google Scholar
- Peña D, Rodríguez J, Tiao GC (2003) Identifying mixtures of regression equations by the sar procedure. In: Bernardo JM, Bayarri MJ, Berger JO, Dawid AP, Herckerman D, Smith AFM (eds) Bayesian Statistics, vol 7. Oxford University Press, Oxford, pp 327–347Google Scholar
- Peña D, Tsay RS, Zamar R (2019) Empirical dynamic quantiles for visualization of high-dimensional time series. Technometrics. https://doi.org/10.1080/00401706.2019.1575285
- Spodarev E (2017) Stochastic geometry, spatial statistics and random fields. Springer, New YorkGoogle Scholar