Quantiles

Here, we'll continue with the Vocab dataset and use stat_quantile() to apply a quantile regression.

Linear regression predicts the mean response from the explanatory variables, quantile regression predicts a quantile response (e.g. the median) from the explanatory variables. Specific quantiles can be specified with the quantiles argument.

Specifying many quantiles and color your models according to year can make plots too busy. We'll explore ways of dealing with this in the next chapter.

Cet exercice fait partie du cours

Intermediate Data Visualization with ggplot2

Afficher le cours

Exercice interactif pratique

Essayez cet exercice en complétant cet exemple de code.

ggplot(Vocab, aes(x = education, y = vocabulary)) +
  geom_jitter(alpha = 0.25) +
  # Add a quantile stat, at 0.05, 0.5, and 0.95
  ___

Modifier et exécuter le code

Cet exercice fait partie du cours

Intermediate Data Visualization with ggplot2

IntermédiaireNiveau de compétence

4.8+

533 reviews

Commencer le cours gratuitement

A picture paints a thousand words, which is why R ggplot2 is such a powerful tool for graphical data analysis. In this chapter, you’ll progress from simply plotting data to applying a variety of statistical methods. These include a variety of linear models, descriptive and inferential statistics (mean, standard deviation and confidence intervals) and custom functions.

Exercise 1: Stats with geoms Exercise 2: Smoothing Exercise 3: Grouping variables Exercise 4: Modifying stat_smooth Exercise 5: Modifying stat_smooth (2)Exercise 6: Stats: sum and quantile Exercise 7: Quantiles

Exercice en cours

Exercise 8: Using stat_sum Exercise 9: Stats outside geoms Exercise 10: Preparations Exercise 11: Using position objects Exercise 12: Plotting variations

The Coordinates layers offer specific and very useful tools for efficiently and accurately communicating data. Here we’ll look at the various ways of effectively using these layers, so you can clearly visualize lognormal datasets, variables with units, and periodic data.

Exercise 1: Coordinates Exercise 2: Zooming In Exercise 3: Aspect ratio I: 1:1 ratios Exercise 4: Aspect ratio II: setting ratios Exercise 5: Expand and clip Exercise 6: Coordinates vs. scales Exercise 7: Log-transforming scales Exercise 8: Adding stats to transformed scales Exercise 9: Double and flipped axes Exercise 10: Useful double axes Exercise 11: Flipping axes I Exercise 12: Flipping axes II Exercise 13: Polar coordinates Exercise 14: Pie charts Exercise 15: Wind rose plots

Facets let you split plots into multiple panes, each displaying subsets of the dataset. Here you'll learn how to wrap facets and arrange them in a grid, as well as providing custom labeling.

Exercise 1: The facets layer Exercise 2: Facet layer basics Exercise 3: Many variables Exercise 4: Formula notation Exercise 5: Facet labels and order Exercise 6: Labeling facets Exercise 7: Setting order Exercise 8: Facet plotting spaces Exercise 9: Variable plotting spaces I: continuous variables Exercise 10: Variable plotting spaces II: categorical variables Exercise 11: Facet wrap & margins Exercise 12: Wrapping for many levels Exercise 13: Margin plots

Now that you have the technical skills to make great visualizations, it’s important that you make them as meaningful as possible. In this chapter, you’ll review three plot types that are commonly discouraged in the data viz community: heat maps, pie charts, and dynamite plots. You’ll learn the pitfalls with these plots and how to avoid making these mistakes yourself.

Exercise 1: Best practices: bar plots Exercise 2: Bar plots: dynamite plots Exercise 3: Bar plots: position dodging Exercise 4: Bar plots: Using aggregated data Exercise 5: Heatmaps use case scenario Exercise 6: Heat maps Exercise 7: Useful heat maps Exercise 8: Heat map alternatives Exercise 9: When good data makes bad plots Exercise 10: Suppression of the origin Exercise 11: Color blindness Exercise 12: Typical problems