Summary

We have looked at two analysis methodologies in this chapter. MBA analysis and segmentation. Both use datasets related to time series, but they transform the data and don't consider the time element explicitly as part of the analysis. In the MBA analysis, we saw that sequence was a sort of proxy for time series. We generated insights by focusing on business problems that were the subject of modeling. MBA lacks the statistical depth and rigor that clustering has. However, neither are strictly statistically-driven analysis scenarios. With MBA, we showcased how it makes intuitive sense to evaluate the association between the products and services offered by a bank and leverage the information. In segmentation via clustering, we showcased how the number of clusters generated differed from the preferred statistical information available to us from two different methodologies...