Inhalt
Kommentar |
Course Description: This course focuses on regression models for cross-sectional as well as time series data. It offers an in-depth discussion of the fundamental multiple linear regression model for cross-sectional data and of methods that aim at modeling the dynamic behavior of economic variables. The latter is especially relevant to quantify the effect of past shocks or changes of variables onto the current or future value of a specific economic variable and to compute forecasts by exploiting this correlation over time. Typical examples are forecasts of macroeconomic variables, such as GDP or inflation, as well as price and volatility forecasts of financial assets, which are essential components for adequate risk management and portfolio allocation decisions. Particular focus will be given on model specification, as well as on the interpretation, consistent estimation, and hypothesis testing of the parameters of the respective models. Methods to assess the fit of the model and its forecast accuracy will also be covered.
Contents:
Regression Analysis: • The Multiple Linear Regression Model • Model Assumptions • Estimation and Inference • Model Diagnostics and Residual Analysis • Parameter Interpretation, Dummy Variables and Interactions of Regressors • The Problem of Endogeneity
Time Series Analysis: • Time Series Data and Stochastic Processes • Autoregressive and Moving Average Models • Seasonalities • Unit Roots and Integrated Processes • Forecasting and Forecast Evaluation • Volatility Models • Introduction to Multivariate Time Series Analysis
|