Problem 2 Regression, models and methods With the data collected by the Dutch National Institute for Coastal and Marine Management (Rijkswaters taat / RIKZ) we are going to study some factors that influence species richness on sandy beaches. The variables included in the RIKZ.txt file are Variable Descripcion Case Number richness numero de especies halladas en una parcela angle2 pendiente de la playa en la parcela NAP altura de la parcela comparada con el nivel medio del mar grainsize diametro medio de los granos de arena humus fraccion de materia organica week en cual de las 4 se tomo la muestra muchas mas variables en el conjunto de datos original (a) First, the richness values must be calculated, which is the number of species found in a plot from the first 75 columns (not counting the first one that can be eliminated). Attention! Note that it is the number of different species, not the number of individuals. Carry out a descriptive study of said variable, especially its distribution. Find a transformation of this variable that improves its properties (you can use the function transform Tuckey() from the rcompanion package). Also check that the transformation log (x + a) with a = 0.1 does not improve normality. (b) Study the regression models: vrichness ~ NAP richness ~ NAP + NAP? Is the term of degree 2 necessary in model 2? Are they comparable linear models? Which seems the most suitable? Draw the two prediction lines on the scatter plot for the variables NAP and richness