We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.OT
< prev | next >
new | recent | 1712

Change to browse by:

stat

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Statistics > Other Statistics

Title: Conducting Highly Principled Data Science: A Statistician's Job and Joy

Authors: Xiao-Li Meng
(Submitted on 2 Dec 2017 (v1), last revised 5 Dec 2017 (this version, v2))
Abstract: Highly Principled Data Science insists on methodologies that are: (1) scientifically justified, (2) statistically principled, and (3) computationally efficient. An astrostatistics collaboration, together with some reminiscences, illustrates the increased roles statisticians can and should play to ensure this trio, and to advance the science of data along the way.
Comments: To appear in the special issue on "The Role of Statistics in the Era of Big Data" in Statistics and Probability Letters (2018)
Subjects: Other Statistics (stat.OT)
MSC classes: 62P99
Cite as: arXiv:1712.00544 [stat.OT]
  (or arXiv:1712.00544v2 [stat.OT] for this version)

Submission history

From: Xiao-Li Meng [view email]
[v1] Sat, 2 Dec 2017 04:13:17 GMT (17kb)
[v2] Tue, 5 Dec 2017 11:52:17 GMT (17kb)
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Link back to: arXiv, form interface, contact.