Often, we want to visualize the distribution of our data to see what values it takes on. Depending on the type of data we have, we may choose to use histograms, kernel density estimates (KDEs), box plots, or empirical cumulative distribution functions (ECDFs). When working with discrete data, histograms are a good place to start.
Let's take a look at the histogram of daily volume traded in Facebook stock:
>>> fb.volume.plot(... kind='hist', ... title='Histogram of Daily Volume Traded in Facebook Stock'... )>>> plt.xlabel('Volume traded') # label the x-axis (discussed in ch 6)
This is a great example of real-world data that is, most definitely, not normally distributed. The volume traded is right skewed, with a long tail to the ...
