In the initial stages of undertaking a regression analysis, the focus is on simple linear regression, which involves the observation of a straightforward relationship between two continuous variables. One of these variables serves as the independent explanatory variable, while the other is designated as the dependent variable, which is intended to be predicted within a model of the form y=am. Subsequently, the multivariable model is employed, where the independent variable, y, is a function of n explanatory variables, xi, also known as regressors. The subsequent step involves incorporating nominal variables into the model. Nominal variables can be integrated into a regression model if it is transformed into n-1 binary variables, where n signifies the total number of categories within the nominal variable. The next step considers the range and distribution of interval variables, in the search of the relationship that exists between the continuous variables and the dependent variable. In this regard, techniques such as marginal value analysis, scaling, and variable transformation are employed. In summary, the most important factors to consider when building a regression model are as follows: