5 Parameters estimation

• Depending on the model type an approriate estimator which gives the “best” parameter estimates should be considered

• Commonly used estimator is the least squares estimator which originates from the OLS method (Ordinary Least Squares)

• Other methods based on OLS have been developed: Weighted Least Squares WLS, Generalized Least Squares GLS, Two Step Least Squares 2SLS, etc.

• In a bivariate model, OLS method is used to find the regression line that best fits the data points by minimizing the sum of the squared residuals; hence the term “least squares”.

• When the OLS method cannot be applied analytically, it is recommended to use an alternative estimator

• An alternative to OLS is the Maximum Likelihood Estimator MLE

• Both OLS and MLE provide the same parameter estimates if the normality assumption is met