Under Multicollinearity A. the OLS estimator cannot be computed. B. two or more of the regressors are highly but not perfectly correlated. C. the OLS estimator is biased. D. there is a perfect linear relationship among two or more covariates.

The correct answer is B. two or more of the regressors are highly but not perfectly correlated.

Explanation:

We say that a multiple regression model presents imperfect multicollinearity when there are strong linear relationships between some of its explanatory variables, which do not become perfect.

Example. - If we try to explain the spending on telephony that families have and use as explanatory variables:

The number of family unit members

Housing spending

The rent

The number of devices with internet connection

Looking at the explanatory variables selected, we will undoubtedly find that they are expected to have high correlations with each other, since the number of devices with internet connection, for example, is undoubtedly related to the number of family unit members. We can also expect that the rent is related to the expenditure on housing and the number of connected devices.

It is very common to find imperfect multicollinearity or simply multicollinearity in econometric models. This would violate the basic hypothesis of independence between the explanatory variables, but already by saying it we said it was difficult to comply.