Course Contents

Non-spherical disturbances, the generalized least squares. Multicollinearity , Ridge Regression, Heteroscedasticity and their solution. Problems of Auto Correlation System of simultaneous linear equations, identification problem, rules of identification, estimation methods,instrumental variables, lagged variables, Dummy Variables.

Course Synopsis

To produce the ability to build regression model in the violation of assumption when ordinary least square method is not appropriate and to learn which method is more preferable in certain given conditions.

Course Learning Outcomes

The learning outcomes associated with course are aimed at students being able to • deal with problem of multicollinearity. • estimate parameters in regression equations when the random disturbances are allowed to be heteroskedastic or auto-correlated • use an appropriate econometric model for a real data problem, interpretation and critical evaluation of the outcomes of empirical analysis • be able to report the results of their work in a non-technical and literate manner