# EViews Course

This free self-paced course is advanced training in time-series economic-data analysis. By the end of the free course, participants should be able to use EViews for modelling and prediction in an expert manner. In each lecture, I show you step-by-step how to solve different forecasting models in EViews.

Throughout the course, participants will get theoretical and hands-on insight into the most prominent time-series data models used for projecting macroeconomic variables, such as the GDP, CPI, Interest rate or any financial assets. We begin with a review of the fundamentals of time series analysis and dynamic models, in the form of linear regression, ARIMA models, ARCH and GARCH models, Vector Autoregression models and more! From basic concepts to advanced time series models! Learn the main econometric methods.

In conclusion, by the end of this course, you will know how to use Eviews for time series analysis. You should now be able to download data, clean it, visualize it, model it, and make forecasts. If you found this course helpful, please show your appreciation by subscribing to my youtube channel, following me on social media, and buying the material to replicate the content.

Thank you for taking the time to go through each of the tutorials!

# 1- How to import Data in Eviews

New to EViews? Learn how to download real economic data from online sources, and upload it to EViews.

# 2- Linear Regression

In this tutorial I teach you how to estimate a simple linear regression and understand the output. We fit a real example using Argentina and Brazil GDP. Finally, I explain why the regression is spurious and what are the signs to detect it.

# 3- Stationarity: Unit Root Tests

Learn about stationarity in EViews. We will do a graph and correlogram analysis. Next, we finish with some formal tests: Augmented Dickey Fuller, Phillips Perron and KPSS test.

# 4- Unit Root Test with Breakpoints

We replicate Perron's paper and learn how to identify structural breaks in our data.

# 5 -HP Filter

Learn how to use the Hodrick-Prescott (HP) filter to decompose a time series into cyclical and trend components. The cyclical component is stationary and will show the percentage deviation from the long run trend.

# 6- ARIMA Models

Learn how to forecast ARIMA models. ARIMA are univariate models, where past information of the variable will help us model how it will behave in the future. We cover the Box-Jenkins 3 step methodology.

# 7- ARIMA Forecast with Confidence Bands

Now that you know how to forecast ARIMA models, let's add confidence bands to the out of sample forecast! Finally, I teach you how to edit the graph to make it look professional.

# 8- ARCH Models

Traditional Econometric models assume that the variance is constant, however, there are periods of high volatility that can influence our variable. Learn about volatility clustering, arch terms, and model the variance of a series. We use Toronto Stock Exchange (TSX) as example.

# 9- ARCH Common Mistakes

Learn what are the three things you should check when estimating ARCH models.

# 10- GARCH Models

GARCH models are an extension of ARCH models. GARCH models tend to be more parsimonious and are a good alternative to high ARCH models. We use Microsoft stock as example.

# 11- Cointegration & Error Correction Model Part 1

Two non-stationary variables can have a long run equilibrium. Learn how to verify so using the Engle and granger method.

# 12- Cointegration & Error Correction Model Part 2

If two series are cointegrated, error correction models will allow us to model the short run model. Let's learn about it!

# 13- VAR Models Part 1

Vector Autoregression (VAR) models, are one of the most popular models for multivariate time series analysis. Learn how to estimate them! In our example, we will replicate Stock and Watson (2001) paper.

# 14- VAR Models Part 2 Impulse Response Functions

Impulse Response Functions (IRFS) show the time path of the variables when they are exposed to a shock in one of the endogenous variables in the model.

# 15- SVAR Models - Long Run Restrictions

SVAR stand for structural vector autoregression models and they imply imposing a restriction on the response matrix based on economic theory. In this case, we will use long run restrictions based on long run money neutrality. We replicate Ender &Lee (1997) paper.

# Datasets to Replicate the Content

1. Linear Regression

2. Stationarity - Unit Root Tests

3. Unit Root Test with Breakpoints

4. Hodrick-Prescott (HP) Filter

5. ARIMA Models

6. ARCH Models

7. GARCH Models

8. Cointegration and Error Correction Model

9. Vector Autoregressions (VAR)

10. SVAR - Long Run Restrictions