Simulation (forecasting and policy analysis).Residual-based (Engle-Granger) co-integration.Violations of the classical assumptions.Principles and methodology of econometric analysis.The course also includes the discussion of the practical implications of employing non-stationary data in estimation, the detection of unit roots in the underlying data-generating processes and concepts of residual-based co-integration and error-correction modelling. Various applications of regression analysis, such as forecasting and policy simulation are discussed and applied. Thereafter, attention is given to the violations of the classical linear regression model that are often encountered in applied econometric work and the consequences of these violations are discussed, as well as practical ways of detecting (diagnostic testing) and solving these problems. The course begins with a basic introduction to the concepts of regression (ordinary least squares estimation) and statistical inference. The aim is to reconcile economic theory with practice, thereby empowering delegates with analytical skills and a hands-on approach to the decision-making processes. This course is of an applied nature and focuses on hands-on experience in estimation, interpretation and evaluation of economic relationships. Presented by the Department of Economics, University of Pretoria 2вЂ“6 November 2020 (5 days) - Pretoria, South Africa

Eviews regression analysis