ABSTRACT
This chapter explores a causal approach to decision theory. The chapter illustrates two case studies related to the subscription fees to and satisfaction with the Italian Radio television RAI, which is the State Company that manages the TV programs in Italy. The first case study describes the application of recursive path models to identify the causes of the number of new subscriptions, whereas the second case study illustrates how structural equation models may be used to assess the TV users’ satisfaction. Common opinion among the statisticians is that causality cannot be investigated through statistical inference or, in other words, statistical inference is not causal inference. Path analysis is usually used to describe the direct dependencies among a set of observable variables, including as special cases models for multiple and multivariate regression, canonical correlation analysis, discriminant analysis, as well as more general models for variance and covariance multivariate analysis.
