Nonlinear Regression
3-4 hours of lectures per week.
Lecturer
Jørgen Granfeldt
Content
In nonlinear regression models the mean of the observations is a nonlinear function of at least one of the parameters. This may create a lot of problems which do not exist in linear models. No closed form expressions for the estimates, the estimates may be biased and the bias may depend on the parametrization. The numerical estimation itself may not be easy. No exact distribution theory for neither estimates nor tests. It is a vast subject with examples from many areas.
Prerequisites
Statistics 1 and Statistics 2.
Literature
To be announced.