A Grammar of Data Analysis

2025/11/04 by Xunmo Yang, Taylor Pospisil, Yang, Xunmo +5 · 1 voice
Mathematics · #Computation (stat.CO) #FOS: Computer and information sciences #stat.CO

paper · pdf · doi:10.48550/arxiv.2511.02156

Abstract

This paper outlines a grammar of data analysis, as distinct from grammars of data manipulation, in which the primitives are metrics and dimensions. We describe a Python implementation of this grammar called Meterstick, which is agnostic to the underlying data source, which may be a DataFrame or a SQL database.

Discussions

• Xunmo Yang, Taylor Pospisil, Omkar Muralidharan, Dennis L. Sun: A Grammar of Data Analysis https://arxiv.org/abs/2511.02156 https://arxiv.org/pdf/2511.02156 https://arxiv.org/html/2511.02156 [bsky, 0 points, 0 comments]

Related

• Data Language Specification via Terminal Attribution
• A Grammar for Reproducible and Painless Extract-Transform-Load Operations on Medium Data
• Declarative Data Analytics: a Survey
• A State-Transition Grammar for Data-Oriented Parsing
• The algebra and machine representation of statistical models