Understanding Relationships in Data

Abstract

By the end of the data gathering and preparation process there should be a single, clean and well formatted data set ready for detailed analysis. The file will contain all predictor variables that are going to be considered for inclusion within the model, as well as the dependent variable (the modelling objective). Potential predictor variables that may have been suggested during the project planning phase, but which have subsequently been shown to contain no meaningful data or data that is obviously incorrect will also have been excluded. For those predictor variables that have been retained, missing and spurious values will have been coded to default values and so on.