Bivariate statistical measures regression and correlation

Bivariate statistical measures regression and correlation

1. Regression

   • In regression analysis is a tool of frequent use in statistics that allows to investigate the relationships between different dependent quantitative variables, the regression analysis is a process or model that analyzes the link between a dependent variable and one or more independent variables
     • Objectives
       • Regression analysis is to construct a function that allows estimating the future value of the study variable.
       • Regression allows you to calculate a conditional average expectation
   • Types
     • Covariates or Independent Variables or Regressor Variables
       • They are used as predictors or are confounding variables that are interesting to control
     • Outcome or Dependent variable or Response variable
       • Attributes on which we want to measure changes or make predictions.

2. correlation

   • Correlation refers to the fact that there is a link between several events. One of the tools that allows us to infer if there is such a link is precisely the correlation analysis
     • Objective
       • Indicate if there is a relationship between the two elements ie variables a little about the nature of said relationship and its strength
       • The correlation can say something about the relationship between the variables, it is used to understand whether the positive or negative relationships the strength of the relationship
   • Types
     • Direct correlation: occurs when increasing one of the variables increases the other. The line corresponding to the point cloud of the distribution is a rising line.
     • Inverse correlation: occurs when increasing one of the variables decreases. The line corresponding to the point cloud of the distribution is a linedecreasing.
     • Null correlation: occurs when there is no dependence of any kind between the variables. In this case the variables are said to be uncorrelated and the point cloud has a rounded shape.