The 2021 EViews Summer School comprises a series of seven 1-day courses running consecutively, with theory and practical applications. All courses will teach econometrics from an applied perspective and demonstrate techniques using EViews 12 software. This edition features many new econometric methodologies which have been introduced in EViews 12. Specifically: long memory dependence in time series models (ARFIMA and FIGARCH and FIEGARCH), second generation panel unit roots tests, machine learning methods for model selection….
The summer school will run for 7 consecutive days, four hours per day, in two sessions per day: morning 10am-12pm and afternoon 2pm-4pm. The morning sessions will discuss theory and the afternoon sessions will offer many practical applications in EViews
Days one and two are strongly recommended to all participants. Participants can then choose to attend all the remaining days, or pick and choose the sessions you would like to attend.
|Morning Session||Afternoon Session||Q&A|
|10am-12pm (London time)||2pm-4pm (London time)||4pm-4:30pm (London time)|
DAY 1, (Prof Trapani): Introduction to EViews basics. The classical linear regression model: representation, estimation, misspecification analysis, forecasting with structural models.
DAY 2, (Prof Trapani): Atheoretical models for stationary series: ARMA models (identification and forecasting); Vector AutoRegressions (identification and forecasting).
DAY 3, (Prof Trapani): Models for nonstationary series: univariate cointegration (representation, estimation and testing); the Vector Error Correction Model (representation, estimation, testing and forecasting).
DAY 4, (Dr. Marchese): Non-linear time series models and forecasting: Markov switching models (EViews syntax, representation, estimation, forecasting); STAR models, volatility models (conditional heteroskedasticity models: estimation, testing), Volatility models (GARCH) and Risk forecasting. Long memory in time series.
DAY 5, (Dr. Marchese): Machine learning methods for model selection, Lasso, Elastic net (EViews syntax, representation, estimation and interpretation). MIDAS models, forecasting and nowcasting (EViews syntax, representation, estimation, interpretation, choice of the weight function). Machine learning methods for MIDAS models.
DAY 6, (Prof Trapani): State Space models: representation (and Eviews syntax); the Kalman filter; examples: time varying parameter models and latent component models; forecasting, filtering and smoothing.
DAY 7, (Prof Trapani): Models for panel data (fixed vs random effects, dynamic panels). Models for secrete choice (taxonomy, estimation, forecasting).
Knowledge of basic econometrics (OLS estimation, mis-specification analysis) is desirable, although not required.
No pre-reading is needed; references for post-course reading will be provided during the course itself.
No previous knowledge of EViews is required.