Tutorials >Concept Tutorials > Decision Trees

Outcomes of decisions depend on both decisions under our control and events, or states of nature, outside our control. One way to represent the relationship between decision options, uncertain events and outcomes is through a decision tree.

(View image of a generic decision tree)

A decision tree can help you map out decision options and uncertain events so that you can see the range of possible outcomes from the choices you could make. Remember that for each potential decision, a range of possible outcomes follows from the uncertain events in question.

As you can see, decision trees represent the decisions, states of nature, and outcomes as discrete--that is, there are only a limited number of options (such as "to buy or not to buy"). In reality, most decisions, states of nature, and outcomes are continuous, rather than discrete. For example, you could "not buy", "buy a little", "buy a lot", or anything in between.

As the climate in the future is uncertain and out of our control (generally), it is represented in decision trees as "uncertain events or states of nature". Climate itself is continuous, but forecasts of climate are generally categorical, so they usually fit well into decision trees because they refer to distinct events: for example, rainfall being either above average, average, or below average; or El Niño or no El Niño.

Above is a very brief description of the concept of decision trees and its qualitative uses, but there is much more to decision trees than that. Their main use is to think explicitly and quantitatively about decision options and uncertain outcomes. But this theory is beyond the scope of this tutorial.