ARIMA-GARCH & ARDL MODELS

SESSION 1 – INTRODUCTION TO FUNDAMENTAL CONCEPTS 

White noise processes; Autocorrelation and Partial Autocorrelation Functions (ACF & PACF)

SESSION 2 – MODEL IDENTIFICATION 

Autoregressive (AR), Moving Average (MA) and ARMA processes

SESSION 3 – PARAMETER ESTIMATION & MODEL DIAGNOSTICS  AIC & BIC; Residual Analysis and Overfitting, ARCH-LM test

SESSION 4 – Case Study I 

More practices on developing ARIMA model, a look into Forecasting

SESSION 5 – VOLATILITY MODELS 

ARCH and GARCH models, Asymmetric GARCH, GARCH-in-Mean

SESSION 6 – MORE EXERCISE ON ARIMA-GARCH-type MODELS 

Hands-on exercise on constructing and comparing ARIMA, GARCH and EGARCH models

SESSION 7 – TIME SERIES REGRESSION 

Autocorrelation: Detection and Remedy, Durbin-Watson statistic

3 softwares for Time Series Analysis: Minitab, SPSS and EViews

SESSION 8 – UNIT ROOT TESTS 

Nonstationarity, Order of Integration, ADF, PP and KPSS tests

SESSION 9 – ENGLE-GRANGER 2-STEP PROCEDURE 

Spurious regression, Stationarity of residuals

SESSION 10 – ARDL COINTEGRATION 

ARDL Bounds Test: Uniform Lagged, Diagnostic Checks: LM and CUSUM tests

SESSION 11 – ARDL LEVEL RELATION 

General to Specific Procedure & Diagnostics, Long-run Coefficients, Short-run EC Representation

SESSION 12 – HANDS-ON ARDL MODELING APPROACH 

More Hands-on exercise on ARDL Modeling