EGY XXX b - Python for Time Series Analysis

Course outline (under review)

Part I: Fundamentals of time series:

1. White noise series, univariate stationary and integrated non-stationary random series

2. Backshift operator, backwards difference operator, and the roots of the characteristic equation of a time series

3. Define a time series through a general linear filter of another stationary random series (particularly of a white noise series)

4. Established time series models: stationary autoregressive (AR), moving average (MA), autoregressive moving average (ARMA) nonstationary integrated (ARIMA) models

5. Random walks with and without drift, particularly those with normally distributed increments

6. A short introduction to multivariate time series models, in particular VAR model

7. Cointegrated processes

8. Estimation, diagnosis and identification of time series models

9. Heteroskedastic (GARCH), non-stationary (e.g. regression with stationary errors) time series models

10. Applications of time series models and forecasts from time series data using Box-Jenkins method and extrapolation

11. Smoothing techniques applied to time series and seasonal adjustment

Part II: Econometric Forecasting:

12. Modelling Trends, Seasonality and Cycles

13. Graphic Method of Forecasting

14. One-step-ahead forecast

15. Point forecast

16. Forecast Interval

17. Vector Autoregressive Model

18. Granger Causality

19. Scenarios Analysis and Impulse Response Function