Choosing the Appropriate Regression from Nine Different Mathematical Models

Choosing the most appropriate mathematical model for establishing a simple regression between two variables which represent two economic indicators or from other scientific fields is an approach that involves many and often quite difficult actions. In the article we rigorously presented the stages to be followed in order to obtain the defining parameters for nine models: linear, power (double logarithmic), exponential, logarithmic, three hyperbolic models (the third is also a Tӧrnqvist-type model) and two exponential-hyperbolic models. The final stage in establishing the most appropriate model is less discussed in the paper. Thus, the simple regression for which the model has the highest Pearson correlation coefficient is chosen as optimal in the article.