Bivariate statistical measures

Bivariate statistical measures

1. Regression

   • We use the word regression to mean the estimation of a variable based on another known value, corresponding to the other variable.

2. Bivariate statistical techniques allow the joint analysis of two characteristics of individuals in a population with the purpose of detecting possible relationships between them. The nature (nominal, ordinal or numerical) of the characteristics under study will determine the most appropriate tools for their analysis.

3. Linear regression

   • Regression variables
     • Covariates
     • Independent variable
     • dependents
     • Response - regressor variables
     • Linear Regression Assumptions
     • Least squares estimators
       • It is a numerical analysis technique framed within the mathematical optimization in which they are ordered.
   • It allows to determine the degree of dependence of the series of values ​​X,Y predicting the value and estimating what would be obtained for a value X that is not in the distribution
     • Simple regression; studies changes in one variable
     • Multiple regression is the possible relationship between several independent and dependent variables.

4. Simple Regression and Correlation Analysis

   • They are frequently used by market researchers to study the relationship between two or more variables.
     • Correlation analysis
     • regression analysis
     • estimation error
     • Correlation coefficient
     • Inferences about the slope coefficient

5. Multiple regression analysis

   • Determine the relationship between independent and dependent variables or predictor and criterion variables
     • It's divided in
       • modified nomenclature
       • Multicollinearity assumption
       • Partial regression coefficient
       • binary variables
       • Variable transformation
       • Correlation coefficient and multiple determination

6. simple linear regression

   • Linear Correlation and Statistical Independence
     • El coeficiente de correlación lineal no hace sino redimensionar el campo de variación de la covarianza entre I y I
     • It is understood as the variance between two variables indicating their degree of linear correlation through the line
   • Regression and simple correlation
     • simple linear regression
       • Estimation of linear regression parameters
       • Regression explains the behavior of a variable, called and explained dependent or endogenous
       • A function is linear in the parameters if they are determined with unit frequency and are not multiplied or divided by any other parameter.

Imagens e arquivos do mapa