In order to perform a hypothesis test to see if X is a signi

In order to perform a hypothesis test to see if X is a significant predictor of Y in simple linear regression, we require five assumptions to hold. Which assumption does not need to be checked graphically?

Solution

There are four principal assumptions which must be check for the use of linear regression models for purposes of inference is:

Linear relationship between dependent and independent variables.

Errors terms hsould be indepedent from each other.

Constant variance (homoscedasticity) of the errors.

The error distribution should be normal.

Hence,

second option does not need to be checked graphically.