Chapter 2 Decision Analysis

2.1 Analytics

Analytics is:

  • the discovery, interpretation, and communication of meaningful patterns in data (Wikipedia)
  • the scientific process of transforming data into insight for making better decisions (INFORMS 2016)

Types of Analytics

  • Descriptive: What is happening?
  • Predictive: What will happen?
  • Prescriptive: What should happen (when we do X)?

2.1.1 Descriptive Analytics

  • Extract information from data, summarise and report
  • Shows the historical and current state of affairs
  • Visualization is important and very valuable

2.1.2 Predictive Analytics

  • Extract information from data and use it to predict classification, trends and behavior patterns.
  • Often the unknown event of interest is in the future, but predictive analytics can be applied to any type of unknown whether it be in the past, present or future.
  • Reduces uncertainty
  • Above a threshold, the accuracy of prediction can be a game changer!

2.1.2.1 Prediction Becomes more Accurate

prediction_becomes_more_accurate

2.1.2.2 Prediction Becomes Cheaper

  • More accurate prediction is now used for traditional prediction tasks (e.g. forecasting weather or demand) and new problems (e.g. translation and navigation).
  • The drop in cost of prediction has two main effects: – creates surprising externalities (e.g. music, video streaming, driving services) – increases the value of complements (data, judgement, and action)
  • If Amazon can predict what consumers want more accurately, they can shift from a shop-then-ship to ship-then-shop strategy.