Slide 6: Running The GAM
To model the distribution of a mammal species of your choice you will initially use a single environmental variable, combined with the distribution of the study species within the country you selected. This data will be divided into a calibration set and an evaluation set, so that you can both train the GAM and verify your results. You will then train the GAM, and evaluate the accuracy of the model. This process will also be repeated, using a set of 3 different variables, and then 5 of the 6 variables, in order to find a model that best describes the distribution of the species in question. You will then use these models to describe the current distribution of the species within your study area, and export this distribution as a dbf file.