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

TL;DR

This paper emphasizes the importance of scientifically justified, statistically principled, and computationally efficient methodologies in data science, highlighting statisticians' crucial roles in advancing data-driven science.

Contribution

It advocates for a principled approach to data science and illustrates statisticians' roles through astrostatistics collaboration and reflections.

Findings

Statisticians can significantly enhance data science practices.

Principled methodologies improve scientific validity and efficiency.

Collaboration in astrostatistics exemplifies these principles.

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.