Answer the following questions briefly:

a) Define the term multicollinearity and explain what effect it can have on multiple regression analysis. How do we detect its presence?

b) Explain which of the following intervals based on the same estimated equations with the fixed values of the explanatory variables would be wider:

i) A 98%prediction interval for an individual value of Y or

ii) A 98% confidence interval for a mean value of Y?

c) Examine the statement, "In multiple regression models, we always use adjusted to choose the best performing model". Explain whether you agree or disagree with this statement.