liveBook · Manning

Cost

As explained in Grokking Bayes

Cost is the expenditure or undesirable consequence incurred when implementing an intervention.

A cost measures how undesirable an outcome is. It can be treated as the negative of utility: higher costs correspond to worse outcomes. In decisions involving interventions, subtract implementation cost from the utility of the result to find net utility. A machine-learning loss function is related, but typically measures mismatch between predictions and observed data. section 12.1

Choosing a policy

A job-training policy with cost 1, expected utility 7.5, and net utility 6.5 compares favorably with a riskier policy costing 2, whose expected utility is 3 and net utility is 1. The second policy’s larger possible payoff does not make it the better choice. section 12.7 section 12.7

Table 12.3 Utility of two available policies under possible scenarios and the corresponding probabilities
Option
Policy A
Policy B
Doing nothing
Cost
1
2
0
Utility
10
5
30
0
0
Probability
0.5
0.5
0.1
0.9
1
Option
Cost
Utility
Probability
Policy A
1
10
0.5
5
0.5
Policy B
2
30
0.1
0
0.9
Doing nothing
0
0
1

Compare the policies’ costs, possible utilities, and outcome probabilities. section 12.7

How Grokking Bayes covers it

How deep it goes
Defines cost, relates it to utility, and subtracts it from utility to calculate net utility.
Examples
Compares website-launch outcomes and job-training policies with different costs, utilities, and probabilities.
What you can do after
Evaluate interventions by subtracting their costs from utility and comparing expected net utility.
How it connects
Connects cost to utility, expected utility, and loss functions.

Also explained in

Compare how 17 books explain cost →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Bayes. The section links point to the parts of the book it draws on.