Multiple regression and non-linear regression analysis.

This chapter introduces the regression techniques for dealing with multiple variables. The multiple linear regression model is described and the testing of the significance, goodness-of-fit and assumptions of multiple linear regression are discussed. The regression analysis of non-linearly related data and the application of the data transformations to 'straighten' the data are shown. The curvilinear regression analysis is also described, as well as the truly non-linear regression analysis: growth models. An example of the application of a multiple linear regression analysis is provided, using the data on potato tuber yield in response to soil nitrogen, phosphate and potassium contents.