Consider an oil-wildcatting problem. A decision maker has mineral rights on a pi
ID: 2497840 • Letter: C
Question
Consider an oil-wildcatting problem. A decision maker has mineral rights on a piece of land that he believes may have oil underground. There is a 30% chance that the decision maker will strike oil if he drills. If he drills and strikes oil, then the net payoff is $180,000. If he drills and does not strike oil, then there will be a $10,000 loss due to the sunk cost. The alternative is not to drill at all, in which case the decision maker's net payoff is $0.
Before the decision maker drill he might consult a geologist who can assess the promise of the piece of land. The geologist can tell the decision maker whether the decision maker's prospects are "good" or "poor". But she (the geologist) is not a perfect predictor. If there is oil, the conditional probability is 0.9 that she will say good. If there is no oil, the conditional probability is 0.85 that she will say poor.
What is the maximum amount that the decision maker (assume he is rational) is willing to pay the geologist for her information? (Calculate EVSI.)
Explanation / Answer
1.
Pay off without geologist engagement = 0.3 * 180000 - 0.7 * 10000 = $61000
2.
Now consider following table to use bayes theorem
There is Oil (30%)
No Oil (70%)
Say Good
0.9
0.1
Say poor
0.15
0.85
Now probability of oil if she say good = 0.9*0.3 / (0.9*0.3 + 0.15*0.7) = 0.72
so probability of no oil if she say good = 1 - 0.72 = 0.28
So pay off if she say good = 0.72 *180000 - 0.28*10000 = $126800
Same way
Now probability of no oil if she say bad = 0.85*0.7 / (0.85*0.7 + 0.1*0.3) = 0.95
so probability of oil if she say good = 1 -0.95 = 0.05
So pay off if she say good = 0.05 *180000 - 0.95*10000 = -$500
Now there is a 50-50 possibility that she will say good or poor
So pay off if we hire geologist = 0.5*126800 - 0.5*500 = $63150
The amount we should be paying is less than 63150 - 61000 = $2150
There is Oil (30%)
No Oil (70%)
Say Good
0.9
0.1
Say poor
0.15
0.85