Question

Which characteristic of a data set makes a linear regression model unreasonable?

Answer 1

Conditions necessary for linear regression :

1.) The regression function is linear with constant but unknown coefficient.

2.) The expected value of the disturbance term is zero.

3.) The disturbance terms have a constant variance.

4.) The disturbance terms are uncorrelated with each other.

A correlation coefficient close to zero makes a linear regression model unreasonable.

Answer 2

Answer:

A correlation coefficient close to 0 makes a linear regression model unreasonable.

Because If the correlation between the two variable is close to zero, we can not expect one variable explaining the variation in other variable. For a linear regression model to be reasonable, the most important thing is to see whether the two variables are correlated. If there is correlation between the two variable, we can think of regression analysis and if there is no correlation between the two variable, it does not make sense to apply regression analysis.

Therefore, if the correlation coefficient is close to zero, the linear regression model would be unreasonable.