Hands-On Data Analysis with Pandas

Binning and thresholds

Sometimes, it's more convenient to work with categories rather than the specific values. A common example is working with ages—most likely, we don't want to look at the data for each age, such as 25 compared to 26; however, we may very well be interested in how the group of 25-34 year-olds compares to the group of 35-44 year-olds. This is called binning or discretizing (going from continuous to discrete); we take our data and place the observations into bins (or buckets) matching the range they fall into. By doing so, we can drastically reduce the number of distinct values our data can take on and make it easier to analyze.

One interesting thing we could do with the volume traded would be to see which days had high ...