2 Carry out a pre-posterior analysis and find the expected value of perfect information (EVPI).

Background to the application of Bayesian decision theory It has been mentioned throughout the chapter that since the 1960s we have seen the development of sophisticated statistical techniques for problem-solving where information is incomplete or uncertain.
The new area of statistics has a variety of names – statistical decision theory, simple decision theory and Bayesian decision theory (after the Reverend Thomas Bayes (1702–61)). These names can be used interchangeably, but for the purposes of this case we use the term Bayesian decision theory.
Bayesian decision theory is a relatively new and somewhat controversial method for dealing with future uncertainties. Applied to forecasting, the technique incorporates the firm’s own guesses as data inputs into the calculation of a sales forecast. There are essentially two ways of conceiving probability:
● as a physical property, inherent to a physical system;
● as a measure of belief in the truth of some statement.