Comparing two continuous columns

Evaluating how two continuous columns relate to one another is the essence of regression. But it goes beyond that. If you have two columns with a high correlation to one another, often, you may drop one of them as a redundant column. In this section, we will look at EDA for pairs of continuous columns.

How to do it…

1. Look at the covariance of the two numbers if they are on the same scale:

   >>> fueleco.city08.cov(fueleco.highway08)
   46.33326023673625
   >>> fueleco.city08.cov(fueleco.comb08)
   47.41994667819079
   >>> fueleco.city08.cov(fueleco.cylinders)
   -5.931560263764761
   

2. Look at the Pearson correlation between the two numbers:

   >>> fueleco.city08.corr(fueleco.highway08)
   0.932494506228495
   >>> fueleco.city08.corr(fueleco.cylinders)
   -0.701654842382788
   

3. Visualize the correlations in a heatmap:

   >>> import seaborn as sns
   >>> fig,...