Course Contents

1. Relaxing the Assumptions of Classical Model 2. Multicollinearity: The nature of multicollinearity, estimation in the presence of perfect and imperfect multicollinearity, consequences of multicollinearity, remedial measures. 3. Heteroscedasticity: The nature of heteroscedasticity, the consequences of heteroscedasticity, the detection of heteroscedasticity, informal method, formal methods (spearman’s rank correlation test, goldfield quandt test), remedial measures. 4. Autocorrelation: Nature, estimation, consequences, Detection: Graphical method, The runs test, Durbin-Watson test, Breusch- Godfrey test, remedial measures. 5. Regression on Dummy Variable: Nature of Dummy Variable, Use of Dummy Variable, ANOVA/ANCOV Models. 6. Forecasting: Forecasting with a single equation model: conditional and unconditional forecasting, variance of forecast error in both cases.

Course Synopsis

The purpose of this course is : to understand how the problem of relaxing the assumption of classical linear regression economics can be tackled. To understand the concept of dummy variables. How to quantify these variables and use in regression models. To build a theoretical base for forecasting. Prerequisite course is Econometrics-I ( Maj/Eco-305)

Course Learning Outcomes

By the end of this course it is expected that the student will: 1. be able to understand the assumptions that underpin the classical regression model 2. be able to quantify the dummy variables and to use these variables in regression analysis. 3. understand the theory of forecasting.