In many situations, standard statistical models may not be suitable for the data being studied; the researcher will need to construct a model that will provide appropriate parameter estimates. This book provides a step-by-step guide to estimating regression models using optimization, maximum likelihood, quadrature and simulation. The text allows readers to develop their own R code for models relevant to the data they are studying. It begins with background material and basic models, building through to more complex situations, such as mixed effects models. It contains data from various disciplines, illustrating how the techniques can be applied to building models for real data problems.