3.3.1 Computing performance-based weights (1)

Course subject(s) Module 3. Performance-based weights and the Decision Maker

During the previous video, you learned about the normalized weights each expert receives based on the calibration and the information score. For N experts, expert has a normalized weight

The normalized weight of an expert is obtained by dividing the combined score  of an expert by sum of all combined scores.

Note that the weights are called normalized because they sum to 1. This will be important in combining experts’ assessments!

Performance-based weights

These weights are called performance-based weights, since they are based on the two scores that objectively evaluate experts’ performance in assessing uncertainty.

“Weights in Grams”
 by biologycorner is licensed under CC BY-NC 2.0


We go back to our Dutch eating habits example.  Recall the calibration and the information scores.

Calibration Score Information score
Expert 1 0.002 1.19
Expert 2 0.608 0.18
Expert 3 0.327 0.46

Recall that the normalized weights should sum to 1.

For this reason, we will further consider w1=0.009, w2=0.417 and w3=0.574.

Creative Commons License
Decision Making Under Uncertainty: Introduction to Structured Expert Judgment by TU Delft OpenCourseWare is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Based on a work at https://online-learning.tudelft.nl/courses/decision-making-under-uncertainty-introduction-to-structured-expert-judgment//.
Back to top