Correlation

A statistical measure called correlation expresses how closely two variables are related linearly. It's a typical technique for describing direct connections without explicitly stating cause and effect. In simple words, it is used to find the relation between two variables.

The correlation value is denoted as r which lies between -1 to +1. If the r = +1 it indicates a perfect positive linear correlation. If the r = -1 it means a perfect Negative linear correlation.

Types of correlation:

• Positive correlation 
• Negative correlation
• Zero correlation or No correlation. 

Let us consider the X and Y variables, that is (Xi, Yi); i = 1,2,3,...,n, if the variables X and Y are potted along the x-axis and y-axis respectively. Then

• if these two variables travel in the same direction (either (X↑, Y↑), (X↓, Y↓)) it is said to be a positive correlation (r > 0).

• if the variables are travels in different directions (either (X↑, Y↓), (X↓, Y↑)) then it is called a negative correlation(r < 0). 

• if the variables X and Y travel linearly (The above two graphs are linear correlation), then it is known as a linear correlation or a non-linear one. For example, Quadratic, cubic, Exponential, logarithmic, etc., 

• if there is no trend or no relationship between the variables then it is called a No correlation or Zero correlation.