Perfect Multicollinearity Case:
Let us first consider the case of perfect multicollinearity, defining the deviations from sample mean:
where are the sample means of the variables respectively. By definition, the coefficient β2 in MLRM is calculated as,
For perfect multicollinearity in our earlier example, we have assumed X3= 1OX2. Let us make it more general and assume X3 = λx2, where λ is a constant as before. This implies that in deviations form we can write x3 = λ2.Substituting this in equition we get,
which is an indeterminate expression. Thus the coefficient β2 is not determinate, i.e., it cannot be estimated. Similarly we can show that the coefficient for β3 is also indeterminate in the presence of perfect multicollinearity.