ABSTRACT
In this chapter the author shows how to make hypothesis tests and construct confidence intervals. These inferential methods are the building blocks for drawing conclusions using models and data. Arguments based on the central limit theorem mean that even if the errors are not normal, inference based on the assumption of normality can be approximately correct provided the sample size is large enough. Unfortunately, it is not possible to say how large the sample has to be or how close to normality the error distribution has to be before the approximation is satisfactory. Permutation tests offer an alternative that needs no assumption of normality. The idea of permutation tests works well in conjunction with the principle of random allocation of units in designed experiments. A sample of convenience is where the data are not collected according to a sampling design. In some cases, it may be reasonable to proceed as if the data were collected using a random mechanism.
