Abstract
This chapter describes the main tool used in the analysis of nominal and ordinal variables—contingency tables and discusses the different measures of association between the variables. Contingency is a fundamental tool in the analysis of nominal and ordinal variables. The result of the analysis of contingency tables is a set of measures of the association between variables. Therefore, in the context of data preparation procedures, one can use these results in the following two areas—(1) to reduce the number of independent variables by removing those that do not show reasonable association with the dependent variable, (2) to compare the distribution of variables in two or more samples to make sure that the model training and validation partitions are not biased with respect to any of their variables. Contingency tables are simply the counts of cross-tabulation of two or more nominal or ordinal variables. The chapter presents the analysis of contingency tables for binary variables and multicategory variables.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have