Input Separability in the Canadian Cable Television Industry: An Application of Generalized Additive Models
We develop an example estimating a non-parametric cost function with Generalized Additive Models (GAMs,). We demonstrate the feasibility of testing for input separability, providing a test for bias in GAIvIs. Our example focuses on cable television (CA TV) provision, and we estimate a new non-parametric cost function for the Canadian industry using financial and operating data collected between 1990 and 1996 This period is of particular importance from a policy perspective because CA TV in Canada was a rate- and entry-regulated industry prior to 1997. The results show that the input separability assumption holds and the GAIvI system provides a good fit to the CA TV data.