In: Statistics and Probability
“Gauss-Markov” assumptions:
1. Data should be randomly sampled from the population.
2. The slope and intercept terms are not correlated.
3. The parameters we are estimating should be linear.
4. Error of the variance is constant.
5. The regressors we are calculating does not have correlation 1.
These assumptions are important as they create the ideal conditions for OLS (Ordinary least square) estimators to be a good estimator. All though in most cases these assumptions are violated still it remains as a good measure of how ideal is the system which we are analyzing.