Question

17. What does linear correlation measure?

18. Which statement regarding linear regression is NOT correct?

a. It is a data model explicitly describes a relationship between predictor and response variables.

b. Linear regression is a linear approach to modeling the relationship between predictor and response variables.

c. Linear regression is performed after correlation analysis

d. Linear regression quantifies the strength of a linear relationship between two variables

Please answer both :)

Thank you I will rate you 5 stars!

Answer 1

Ans 17.

Linear Correlation is basically a way of getting to know how 2 variables are related to each other in our model.

This correlation exists between -1 and +1 , where 0 means no straight line (in the plot).

More the increase in Y with increase in X , more closer the value is to 1.

More the decrease in Y with increase in X , more closer the value is to -1.

0 correlation means there is no correlation.

As mentioned above, there is positive correlation as well as negative correlation.

Example of Positive correlation:

Let's assume, variable Y is for Salary and variable X is Years of experience. We all know that the general trend is that salary increases with increase in experience, so To get a idea about how our plot will look for positive correlation, you can check the plot example below.

Example of Negative correlation:

Let's assume, variable Y is Bank Balance  and variable X is expenditure. We all know that the general trend is that Bank Balance decreases with increase in expenditure, so To get a idea about how our plot will look for negative correlation, you can check the plot example below.

Ans 18.

d. Linear regression quantifies the strength of a linear relationship between two variables.

The above statement is not really true because Linear regression is mainly to predict Y from X and it is actually the correlation that quantifies the strength of a linear relationship between 2 variables.