Anscombe’s quartet

As explained in Delivering Machine Learning Projects: From design to deployment

Anscombe’s Quartet is a set of four paired data sets with nearly identical summary statistics but substantially different distributions and relationships.

Anscombe’s quartet contains four paired data sets with different joint properties but identical-looking summary values: mean xₙ = 9, mean yₙ = 7.5, and Pearson correlation = 0.816. Plotting the data reveals relationships that summary statistics conceal. section 6.1.3

EDA checkpoint

Add a visualization ticket before modeling: plot the full data set, inspect how points relate, and compare the visible distributions with the reported summaries. section 6.1.3

Figure 6.1 Anscombe’s Quartet. Anscombe.svg: Schutz (label using subscripts): Avenue, CC BY-SA 3.0 (https://creativecommons.org/licenses/by-sa/3.0), via Wikimedia Commons

Use the quartet figure to compare four plots whose shared summaries obscure their different distributions and relationships. section 6.1.3

How Delivering Machine Learning Projects: From design to deployment covers it

How deep it goes

Defines the deceptive-statistics problem and motivates plotting, without developing a detailed statistical investigation of every data-set property.

Examples

Uses four paired data sets sharing means and Pearson correlation while displaying different joint distributions and relationships.

What it assumes

Assumes familiarity with summary statistics and correlation while reinforcing why visual exploratory analysis is necessary.

What you can do after

Evaluate/understand when plots expose structure that summary statistics hide before designing data sets for modeling.

How it connects

Connects summary statistics with exploratory data analysis, scatter plots, data visualization, and Pearson correlation.

liveBookmanning.com

Read Delivering Machine Learning Projects: From design to deployment on liveBook Get the book at manning.com

Related concepts

• descriptive statisticscontrasts with · Descriptive statistics are methods for summarizing and characterizing observed data without drawing conclusions beyond those data.

This explainer was generated by Manning's knowledge graph from the text of Delivering Machine Learning Projects: From design to deployment. The section links point to the parts of the book it draws on.