ABSTRACT
This chapter shows the evolution of statistical modelling from the linear model (LM) through the generalized linear model (GLM) and the generalized additive model (GAM) to the generalized additive model for location, scale and shape (GAMLSS). It introduces GAMLSS and focuses on the GAMLSS model using ideas from its predecessors, in particular from linear regression models, generalized linear models and generalized additive models. The chapter utilizes a relatively simple example; the Munich rent data, to demonstrate why GAMLSS is needed. The rent data come from a survey conducted in April 1993 by Infratest Sozialforschung, where a random sample of accommodation with new tenancy agreements or increases of rents within the last four years in Munich was selected, including single rooms, small apartments, flats and two-family houses. The median rent increases if the flat has central heating, and increases as the location changes from below average to average and then to above average.
