The main stories are "Diagnostics on Large Data Sets": a section from the updated GARCH course which explains the common problem of models on large data sets (1000's of observations) failing to pass standard diagnostics even when the model seems perfectly fine and "Toda-Yamamoto Causality Test: A Cautionary Tale" which explains how the often-used alternative to the Granger test is fundamentally flawed.

The ARCH/GARCH and Volatility Models e-course has been updated to a 2nd edition, which includes new and expanded treatment of many topics in GARCH modeling.

The latest build of 9.2 is now available for all three platforms. The "e" build dramatically simplifies the process for customizing graphs styles, adds new graphics exports to Windows, and generally runs faster than previous releases. 9.20 also includes quite a few enhancements to the GARCH instruction, improvements to the Markov Switching support routines, two new and powerful methods for non-linear estimation and much more. See RATS 9.2 for more information. This includes information about changes with 9.1 and 9.0 if are using an older version.

The main stories are "Markov-Switching GARCH models", which is a summary of the section from the new edition of the Structural Breaks course; "How to Switch if you Must" describes the differences between three common types of regime-based behavior (structural break, threshold break and Markov switching) and how to choose which is appropriate; "Evaluation of GARCH Forecasts" looks at difficulties with using common forecast error statistics (like RMSE) in evaluating out-of-sample behavior of GARCH models.

The Structural Breaks and Switching Models e-course has been updated to a 2nd edition, which includes new and expanded treatment of many topics in models with thresholds, breaks and Markov switching.

The State-Space/DSGE e-course has been updated to a 2nd edition, which more than doubles the size of the original.

We've posted the worked examples for the 4th edition of Walter Enders' *Applied Econometric Time Series*, Wiley, 2015. This is an intermediate
book on applied time series, and covers a broad range of applications from ARIMA models to GARCH models to cointegration.
See the single book-browser for more.

The VAR e-course has been updated to a 2nd edition, which adds over 50% more material.

We've posted the worked examples from Martin, Hurn and Harris, *Econometric Modelling with Time Series: Specification, Estimation and Testing,*
Cambridge University Press, 2013. This is a fairly advanced book which looks at time series analysis primarily by means of the likelihood principle.
See the single book-browser for more.

The long-awaited full version of our Handbook to accompany Juselius' *The Cointegrated VAR Model* text is now available!

Estima develops and sells RATS (Regression Analysis of Time Series), a leading econometrics and time-series analysis software package.

RATS is used worldwide by economists and others for analyzing time series and cross sectional data, developing and estimating econometric models, forecasting, and much more.

RATS Product Information

Place a credit card order

You can use the following links to learn more about our other products:

- CATS 2.0 (Cointegration Analysis Procedure, requires RATS)
- Web courses and seminars
- Economic Databases
- Econometric Textbooks
- Windows TD: a chess tournament pairing program.