Nonlinear regression tools give us the ability to specify an expectation function that goes beyond the Linear, Quadratic, or Cubic models. Applications of nonlinear regression are present where the Linear models fail to fit very well. Models for exponential growth and decay rates are good examples, where the nonlinear tools will provide a better fit to the data. The use of nonlinear regression tools is more complicated than the Linear models, and initial estimates of the coefficients must be provided.

Here we will use the data from the Oxford weather station to define our own expectation function, set the initial parameters of the function, and find the parameters of the coefficients. We will concentrate on the results from 2000 onwards, hence the initial steps will be to subset the worksheet.

From these results, we will define a regression model to predict the mean maximum monthly temperature from the month of the year.