why we use correlation measuring association

[kartikey agarwal]

it will  be a great pleasure for me if you can tell me why we use correlation measuring association? what its use.

[statistics-1]

Hi,

 Consider the dataset of paired values (x1, y1), (x2, y2), . . . , (xn, yn). Correlation coefficient measures the degree to which larger x values go with larger y values and smaller x values go with smaller y values.  For more clarification please go through the lectures again.

Thanks 

Ankur Upadhyay

Course Support Team

[Sanjaykumar]

Correlation is a measure of linear association between two numerical variables. 

i.e. if x and y are two variables, then we have to know cause and effect relationship between them.

Does change in x implies change in y? What will be the extent of this change?

Answers to these and many other questions are given using this measure.

For example.

For correlation r =1, => there is strong association between x and y and change in x affects change in y in same direction.

                            r=-1 => there is strong association between x and y and change in x affects change in y in opposite direction.

In lecture, ma'am has already told about this and other values of r.

Real life situations may include

Effect Chane in temperature and pressure
Effect of change in price of some object over given years.