LAB: Univariate analysis

Published

March 18, 2025

M1 MIDS/MFA/LOGOS

Université Paris Cité
Année 2024

Course Homepage

Moodle

Univariate numerical samples

Code
to_be_loaded <- c("tidyverse", 
                  "magrittr",
                  "skimr",
                  "lobstr"
)

for (pck in to_be_loaded) {
  if (!require(pck, character.only = T)) {
    pak::pkg_install(pck) # ,,    repos="http://cran.rstudio.com/")
    stopifnot(require(pck, character.only = T))
  }  
}

Objectives

In Exploratory analysis of tabular data, univariate analysis is the first step. It consists in exploring, summarizing, visualizing columns of a dataset.

In common circumstances, table wrangling is a prerequisite.

Then, univariate techniques depend on the kind of columns we are facing.

For numerical samples/columns, to name a few:

  • Boxplots
  • Histograms
  • Density plots
  • CDF
  • Quantile functions
  • Miscellanea

For categorical samples/columns, we have:

  • Bar plots
  • Column plots

Dataset

Since 1948, the US Census Bureau carries out a monthly Current Population Survey, collecting data concerning residents aged above 15 from \(150 000\) households. This survey is one of the most important sources of information concerning the american workforce. Data reported in file Recensement.txt originate from the 2012 census.

In this lab, we investigate the numerical colums of the dataset.

After downloading, dataset Recensement can be found in file Recensement.csv.

Choose a loading function for the format. Rstudio IDE provides a valuable helper.

Load the data into the session environment and call it df.

Table wrangling

Question

Which columns should be considered as categorical/factor?

Coerce the relevant columns as factors.

Search for missing data (optional)

Question

Check whether some columns contain missing data (use is.na).

Useful functions:

Analysis of column AGE

Numerical summary

Use skimr::skim()

Question

Compare mean and median, sd and IQR.

Are mean and median systematically related?

Question

Are standard deviation and IQR systematically related ?

Boxplots

Question

Draw a boxplot of the Age distribution

Question

How would you get rid of the useless ticks on the x-axis?

Histograms

Question

Plot a histogram of the empirical distribution of the AGE column

Question

Try different values for the bins parameter of geom_histogram()

Density estimates

Question

Plot a density estimate of the AGE column (use stat_density.

Question

Play with parameters bw, kernel and adjust.

Question

Overlay the two plots (histogram and density).

ECDF

Question

Plot the Empirical CDF of the AGE distribution

Question

Can you read the quartiles from the ECDF pplot?

Quantile function

Question

Plot the quantile function of the AGE distribution.

Repeat the analysis for SAL_HOR

Question

How could you comply with the DRY principle ?