Decision making frameworks

A decision tree is a chronological representation of the decision process, which, as you know, can be very complex. A decision tree is made of two types of nodes.

1. Decision (choice) nodes: squares used to represent decisions
2. Chance (‘states of nature’) nodes: circles used to represent ‘states of nature’/chance-dependent points. (Note: you need to ensure that the probabilities along any outgoing branch sums to one.)

While individual decision trees can get incredibly complex, the steps for building them are the same.

1. Construct the decision tree using squares to represent decisions and circles to represent uncertainty.
2. Evaluate the decision tree to make sure all possible outcomes are included.
3. Calculate the tree values working from the right side towards the left.
4. Calculate the values of uncertain outcome nodes by multiplying the value for the outcomes by their probability (ie the expected value).

They use similar shapes and both show relationships, but what are some of the differences between influence diagrams and decision trees?

At a high level:

• Influence diagrams provide basic information and are more appealing to present
• Decision trees provide more complex information but are less appealing to present.

What other differences do you think there are between influence diagrams and decision trees? We outlined some of the advantages and disadvantages of influence diagrams, but what are the pros and cons of decision trees?

Share your thoughts in the comments. In the next step, you will be able to practise what you’ve learned here by creating a decision tree from a realistic scenario.