Approximate confidence regions for parameter subsets in a non-linear model

Rok, strany: 2003, 227 - 236

Using a cubic approximation to the solution locus instead of the usual linear or quadratic approximations, improved approximate confidence regions for parameter subsets in nonlinear regression models are derived. Examples are given which show that these regions have true coverage closer to the nominal level than regions currently used.

ISO 690:

Potocký, R. 2003. Approximate confidence regions for parameter subsets in a non-linear model. In Tatra Mountains Mathematical Publications, vol. 26, no.2, pp. 227-236. 1210-3195.

APA:

Potocký, R. (2003). Approximate confidence regions for parameter subsets in a non-linear model. Tatra Mountains Mathematical Publications, 26(2), 227-236. 1210-3195.