Please enable JavaScript.
Coggle requires JavaScript to display documents.
Behavioral eco-lect 1: Introduction part 3 (Choice under certainty…
Behavioral eco-lect 1: Introduction part 3
Choice under certainty
Ordinal and cardinal utility
Assume that utility function u(.) represents preferences ≽ .
Suppose we construct a new utility function v(.).
Consider preferences ≽ over consumption bundles.
We construct it as follows: for every x, we set v(x) = 4u(x) + 10.
See slide 31-33/53
Utility function v defined by v(x) = 4u(x) + 10 (for all x), represents the same preferences as u.
Suppose Bill has utility function v and derives 18 units of utility from an apple.
Then we cannot say that Bill likes apples more than Ann does!
Suppose Ann has utility function u and derives 2 units of utility from an apple.
Utility is used to compare consumption bundles within one person,
and not to compare satisfaction between different people.
Whether we use u or v to model someone’s preferences, both will predict the same choices.
Utility u is
cardinal
if we can replace it with utility v with v(x) = f(u(x)) (for all x) as long as f is a linear increasing function (e.g. f(u) = 4u + 10)
Utility u is
ordinal
if we can replace it with utility v with v(x) = f(u(x)) (for all x) as long as f is an increasing function (e.g. f(u) = 4u + 10, or f(u) = eu)
If utility is
ordinal
, utility levels have no meaning other than to order consumption bundles from best to worst in terms of preferences.
cardinal
, utility levels have no meaning other than
(1) to order consumption bundles from best to worst in terms of preferences.
(2) to say how utility differences compare to each other (e.g. I find an increase from €90 to €100 a better improvement than an increase from €100 to €110)
Note: (1) and (2) both within the same person.
What is utility?
Utility as we use it in this course and as described in the previous slides is also called
decision utility
– it is used to predict choices.
The
experienced utility
of an outcome is the measure of the hedonic experience of that outcome (Kahneman, 1994)
Decision utility can be different from experienced utility
Maximization of decision utility may be different from maximization of experienced utility.