learning by doing Ralph faees a ehoice problem with three equally likely states. You rnight think of these as reflecting the state of the econorny or the produet market' s acceptanee of a new produet design. Four possible acts are available, with the following eash outeorne possibilities:

state 1 state 2 state 3 aet 1 0 0 0 2 0 0 100 3 -9,000 -9,000 900 4 100 100 100 The eateh is Ralph must (i) pick one of the four aets, (ii) observe the eash outeorne, and (iii) pick from among the four aets again. Whatever state obtains is constant across the choiees. Thus, if aet 2 is pieked and an outeorne of 100 rnaterializes, Ralph knows state 3 is present in the subsequent choiee.

aJ Assurne Ralph is risk neutraI and seeks to rnaxirnize the sum of the expeeted eash flow. Detennine Ralph's optimal behavior.

b] How does Ralph use the initial outeorne in your solution? Is it rnerely a pile of eash or does it eonvey infonnation as weIl?

AppendixLO1