stats_visualization/

Visualizing Distributions

Although summary statistics are very useful and should be utilized to describe data, visualizing distributions of data can provide very quick intuition about how to proceed in an analysis. Tools like histograms and box plots provide a visual representation about the distribution of variables.

Example Histogram

from here

Example Box Plot

from here

Notes

• A histogram is an estimate of the probability distribution of a continuous variable created by "binning" data together into a certain number of bins defined by variable ranges.
• A box plot also displays the distribution of data. In this case, based on a five number summary: minimum, first quartile, median, third quartile, and maximum

Links

Stat Trek
Khan Academy - Statistics
Bayesian Statistics
Elements of Statistical Learning

Code Review

github.com/gonum/plot docs
Histogram of a normal distribution
Histograms with real data
Box Plot with various distribution
Box Plots with real data

Exercises

Exercise 1

Part A Create a box plot of the values in the third column bmi of diabetes.csv.

Template | Answer

Part B By examining the box plot, draw (on paper) what you think a histogram of the bmi values would look like.

Exercise 2

Create a histogram of the values in the third column bmi of diabetes.csv. Compare the histogram to your drawing created in Part A.

Template | Answer

All material is licensed under the Apache License Version 2.0, January 2004.