liveBook · Manning

Opportunity loss

As explained in Timeless Algorithms, The Foundational Ideas

Opportunity loss is the loss incurred by choosing an action instead of the best action that would have been selected with the relevant information.

A decision made before demand is known can be judged by what each choice sacrifices in each possible state. Opportunity loss measures the utility forgone by choosing an act other than the one that would have been best if the true state were known. section 7.3

L(a, θ)=U(a*, θ)-U(a, θ)

Here, a*(θ) is the optimal act for state θ, and a is the chosen act. An optimal choice has zero opportunity loss; an inferior choice has positive loss. section 7.3.1

Expected opportunity loss is the probability-weighted average of losses across possible states. Minimizing it is mathematically equivalent to maximizing expected utility: one emphasizes sacrificed satisfaction, the other expected satisfaction. section 7.3 section 7.3.1

Table 7.6 Utilities and opportunity losses for the product launch decision.
Act
High demand utility
Low demand utility
Expected utility
Expected opportunity loss
Launch
+100
-40
(0.6)(100) +     
(0.4)(-40) = +44
(0.6)(0) +  
(0.4)(60) = +24
Delay
+70
+20
(0.6)(70) + 
(0.4)(20) = +50
(0.6)(30) + 
(0.4)(0) = +18

For the launch decision, delaying has expected opportunity loss 18, compared with 24 for launching; it also has the higher expected utility, 50 versus 44. section 7.3.2

How Timeless Algorithms, The Foundational Ideas covers it

How deep it goes
Defines state-specific loss, probability-weighted expected loss, and its equivalence to maximizing expected utility.
Examples
Compares launching and delaying under high and low demand, with utilities and expected losses for each act.
What you can do after
Evaluate uncertain acts by minimizing expected opportunity loss and identify the same preferred act as expected utility maximization.
How it connects
Connects opportunity loss with expected utility, regret, and probabilities over states of nature.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Timeless Algorithms, The Foundational Ideas. The section links point to the parts of the book it draws on.