BIVARIANT STATISTICAL MEASURES

BIVARIANT STATISTICAL MEASURES

1. Allow the joint analysis of two characteristics of the individuals of a population with the purpose of detecting possible relationships between them.

   • Very often it is found in practice that a relationship exists between two (or more) variables. For example: the weights of adult men depend somewhat on their heights; the lengths of the circumferences and the areas of the circles depend on the radius, and the pressure of a mass of gas depends on its temperature and its volume.
     • If all the values of the variables meet exactly an exact relationship, then it is said that the variables are perfectly correlated or that there is a perfect correlation between them or, more simply, that there is a function or a formula that relates them. Thus the length L of a circle and its radius r are perfectly correlated since it is exactly verified that: L = 2pr
       • On the contrary, if two dice are thrown simultaneously a few times, there will be no relationship between the points obtained on each die (unless the dice are loaded), that is, there will be no correlation between the scores of each die. In other cases, there seems to be some correlation, even if it is not perfect. For example, the variables height and weight of individuals seem to have a certain degree of relationship even though there is no formula that allows us to guess the weight of an individual given their height.

2. Regression statistical measure

   • In statistics, regression analysis is
     • It is a statistical process for estimating relationships between variables. Includes many techniques for multi-variable modeling and analysis, when attention is focused on the relationship between a dependent variable and one or more independent variables

3. Statistical measure correlation

   • In statistics, correlation analysis is
     • Correlation is a statistical measure that expresses the extent to which two variables are linearly related (that is, they change together at a constant rate). It is a common tool for describing simple relationships without making statements about cause and effect.