George's annual Monkey show has been scheduled for Los Angeles on August 15. The profits obtained in putting on a Monkey show are heavily dependent on the weather. In particular, if the weather is rainy, the show loses $15,000. If the weather is sunny, the show makes a profit of $10,000. All days are assumed to be either rainy or sunny. George has already put a $1,000 deposit down on the site, to reserve it and this deposit (which is a separate matter from the previously mentioned possible loss or profit) will not get refunded whether the monkey show occurs or not. Historical data show that rain has occured on 25% of all August 15th dates over the last 100 years. To help in his decision making, George could get a custom forecast of the Los Angeles weather from Peggy's forecast, a pay-by-the-forecast business. Peggy's forecasting accuracy is as follows: on days where it does rain, Peggy has correctly forecast this weather 90% of the time; but on days where it's sunny, Peggy has correctly forecast this weather only 80% of the time.
Provide the following:
1. Ignoring entirely the possibility of buying a weather forecast form Peggy, what is the expected value of perfect information (EVPI) in the decision scenario? (You must provide a properly drawn and analyzed ("folded back") decision tree as a part of your answer.
2. What is the maximum amount George should be willing to pay Peggy for a forecast? This is, the EVSI?