Bivariate statistical measures of regression and correlation

Più simili a questo

Transform

da J Phillips

Chapters 12, 13, 14, & 15

da Jessica Webster-Craig

Data Mining

da Merik LS

Monte Carlo Analysis in Corporate Finance

da Carlos Bellon

Bivariate statistical measures of regression and correlation

Regression

Variable Types

Dependent variables:

are the attributes on which we want to measure changes or make predictions.

Independent variable:

they represent inputs or causes, that is, potential reasons for variation.

The regression allows to calculate a conditional (average) expectation.

Build a function that allows estimating the future value of the study variable.

Regression analysis is a process that analyzes the link between a dependent variable and one or more independent variables.

It is a tool frequently used in statistics, which allows to investigate the relationships between different quantitative dependent variables.

Correlation

degrees of correlation

null correlation:

here there is no type of pattern or relationship between them.

weak correlation:

it will be weak the further apart the points are on the line.

strong correlation:

the correlation will be strong the closer the points are to the line.

Correlation coefficient

It is a measure that allows knowing the degree of linear association between two quantitative variables (X, Y).

Correlation Types

Null correlation:

occurs when there is no dependency of any kind between the variables. In this case it is said that the variables are uncorrelated and the cloud of points has a rounded shape.

Inverse correlation:

occurs when increasing one of the variables, the other decreases. The line corresponding to the cloud of points of the distribution is a decreasing line.

Direct correlation:

occurs when increasing one of the variables, the other increases. The line corresponding to the cloud of points of the distribution is a growing line.

Goal

Tell us if there is a relationship between two (2) events. That is, a little about the nature of the relationships between variables.

It is any statistical relationship, causal or not, between two random variables or bivariate data.