Question

3) List all 7 of the Gauss-Markov conditions as they are listed in your textbook. Next to each condition, write one sentence translating its meaning into commonsense English- what does this condition mean to you?. (Total of 14 sentences).

Answer 1

1 - Linearity - This means that the model is linear in coefficient and error terms. This assumption addresses the functional form of the model.

2 - The error term has a population mean of 0. This error term accounts for variation in the dependent variable that the independent variables are not able to explain.

3 - All the independent variables are uncorrelated with the error term. If an independent variable is correlated with the error term, then we can use the independent variable to find out the error term which contradicts the assumption 2 written above. This assumption is also known as exogeneity.

4 - Observations of the error term are uncorrelated with each other. One observation should not be able to predict the next observation. This problem is known as Autocorrelation.

5 - The error term has constant variance. This assumption means that there is no heteroscedasticity. In other words, we can say that the variance doesn't change for each observation or over a range of observations.

6 - No independent variable is a perfect linear function of other independent variables. This condition is known as multicollinearity. OLS is not able to distinguish one variable from another if they are perfectly correlated.

7 - The error term is normally distributed. This assumption (if true) produces better confidence intervals but is not a necessary one. This can be asses through a normal probability plot.