BIVARIATE STATISTICAL MEASURES

BIVARIATE STATISTICAL MEASURES

1. REGRESSION

   • It is a frequently used tool in statistics.
     • It allows
       • investigate relationships between different quantitative variables
         • types
           • Covariates or independent variables
             • are confounding variables that we want to control
           • dependent variable or response variable
             • are attributes we want to measure
   • its objective
     • is to build a function that allows estimating the future value of the study variable
   • multiple regression
     • can be linear and curvilinear
       • regression coefficient
         • can be positive, negative or null
         • indicates the number of units that the dependent variable "y" is modified by the effect of the change of the variable independently "X" or vice versa is a unit of measure

2. correlation

   • refers to the fact that there is a link between several events
     • It allows us to infer if there is such a link is precisely the correlation analysis
       • types
         • direct correlation
           • occurs when increasing one of the variables increases the other
         • inverse correlation
           • occurs when one variable increases and the other decreases
         • null correlation
           • occurs when there is no dependency of any kind between the variables
   • its objective
     • is used to understand: whether the relationship is positive or negative

3. Dispersion diagram

   • is defined as the degree of distance of a set of values ​​from their mean value

4. coefficient of determination r2

   • It is a statistical measure where is the proportion of the total variance of the variable explained by the regression
     • is dominated R squared and serves to reflect the goodness of fit of a model to the variable that is intended to explain

Bilder und Dateien der Mindmap