Linear correlation and regression

Correlation is a statement showing the close relationship between the 2 (two) or more variables and the relationship between these variables based on the results of scientific research.
In this article we correlations is limited to two variables, which we will later use the symbol variables x and y.
The first known variable is called the independent variable or variables that affect, while others, who have not known is called the dependent variable or variables that are affected.
To study the relationship two variables can be done with two observations, the observation of the scattering diagram and observations on a chart or graph.
1. Observation of the scattering diagramFirst, two variables that have a relationship, the variables x and y is described in a scattered diagram. The next step we interpret the nature of the relationship based on the chart. In general, the nature of the relationship variables x and y can be classified into 3 groups:a. A positive relationship, if the variable x goes up or down, then the variable y goes up or down.b. Negative relationship, if the variable x goes up or down, then the variable y down or up.c. There is no relationship. If the relationship between the variables x and y can not be described by a straight line (linearly).
2. Observation of the chartsWith this method the two variables x and y respectively depicted in the diagram. If the second graph shows the same direction, there is a relationship between two variables. If the two graphs did not show the same direction, meaning that there is no relationship between the variables x and y.
RegressionIn statistical regression, calculated how big the role or contribution of the independent variables in shaping the dependent variable. While other variables are not taken into account as required variables as constants.
The regression equation:Y = a + b. x
Where:Y = the dependent variable or variables causedx = the independent variable or causeb = coefficient of xa = constant.
The method used1. Free methods2. The method of least squares sum.