In this section of the blog, we provide
a summary of the Add-ins that have been
released or updated within the previous few months, and we announce the winner
of our “Add-in of the Quarter” prize!
As a reminder, EViews Add-ins are additions to the EViews interface or command language written by our users or the EViews Development Team and released to the public. You can install Add-ins to your EViews by using the Add-ins menu from within EViews, or by visiting our Add-ins webpage.
As a reminder, EViews Add-ins are additions to the EViews interface or command language written by our users or the EViews Development Team and released to the public. You can install Add-ins to your EViews by using the Add-ins menu from within EViews, or by visiting our Add-ins webpage.
We have five new Add-ins within the last few months:
- BFAVAR
- SRVAR
- TVSVAR
- FORCOMB
- TSCVAL
BFAVAR
The BFAVAR Add-in, written by Davaajargal Luvsannyam, estimates Factor Augmented Vector Auto Regression (FAVAR) models using the one-step Bayesian likelihood approach.
Unlike the FVAR Add-in, which takes the two-step principle component approach to the FVAR model estimation, the BFAVAR Add-in takes a Bayesian perspective, treating the model parameters as random variable. The Add-in implements the multi-move Gibbs sampling explained in Bernanke, Boivin, and Eliasz (2005)1. It is known that the Bayesian likelihood-based estimation based on MCMC methods comes at the cost of computation burden.
The BFAVAR Add-in, written by Davaajargal Luvsannyam, estimates Factor Augmented Vector Auto Regression (FAVAR) models using the one-step Bayesian likelihood approach.
Unlike the FVAR Add-in, which takes the two-step principle component approach to the FVAR model estimation, the BFAVAR Add-in takes a Bayesian perspective, treating the model parameters as random variable. The Add-in implements the multi-move Gibbs sampling explained in Bernanke, Boivin, and Eliasz (2005)1. It is known that the Bayesian likelihood-based estimation based on MCMC methods comes at the cost of computation burden.
SRVAR
The SRVAR Add-in, also written by Davaajargal Luvsannyam, performs analysis of Bayesian Sign Restricted Vector Auto Regression (SRVAR) models using the flat Normal-inverse Wishart prior.
There is fast growing literature that identifies structural shocks by imposing sign restrictions on the responses of (a subset of) the endogenous variables to a particular structural shock. The SRVAR Add-in employs the Uhlig (2005) rejection method2 to identify structural shocks. This Add-in helps us pin down the impact of structural shocks of the model recursively using the Cholesky decomposition in the process of Bayesian MCMC calculations.
The SRVAR Add-in, also written by Davaajargal Luvsannyam, performs analysis of Bayesian Sign Restricted Vector Auto Regression (SRVAR) models using the flat Normal-inverse Wishart prior.
There is fast growing literature that identifies structural shocks by imposing sign restrictions on the responses of (a subset of) the endogenous variables to a particular structural shock. The SRVAR Add-in employs the Uhlig (2005) rejection method2 to identify structural shocks. This Add-in helps us pin down the impact of structural shocks of the model recursively using the Cholesky decomposition in the process of Bayesian MCMC calculations.
TVSVAR
The TVSVAR Add-in, again written by Davaajargal Luvsannyam, performs Bayesian analysis of Time Varying Structural Auto Regression (TVSVAR) models introduced in Primiceri (2005)3.
A common assumption in the VAR model analysis is that the VAR coefficients are constant over time. However, in many applications, it may be more appropriate to consider time variations in the VAR coefficients. Following Primiceri, this Add-in implements the structural VAR model which allows for both stochastic volatility and time-varying regression parameters.
FORCOMB
The FORCOMB Add-in, written by Yongchen Zhao, provides a way to combine multiple candidate forecasts into a robust real-time forecast.
Time series forecasting is a continuously growing research area in many domains of business, finance, engineering and demography, etc. Improvements to the accuracy of forecasting have received extensive attention from researchers. In different publications, it has been observed that combining multiple forecasts improved the overall forecast accuracy. This Add-in provides different types of robust (weighted) forecast combination techniques such as S-After, L-After, h-After, L210-After, Scancetta (2010) MLS4, simple average, trimmed mean, winsorized mean and Bates-Granger (1969)5 methods.
Time series forecasting is a continuously growing research area in many domains of business, finance, engineering and demography, etc. Improvements to the accuracy of forecasting have received extensive attention from researchers. In different publications, it has been observed that combining multiple forecasts improved the overall forecast accuracy. This Add-in provides different types of robust (weighted) forecast combination techniques such as S-After, L-After, h-After, L210-After, Scancetta (2010) MLS4, simple average, trimmed mean, winsorized mean and Bates-Granger (1969)5 methods.
TSCVAL
The TSVCAL Add-in, written by James Lamb and Rita Linets, performs rolling estimation and out-of-sample forecast evaluation from EViews’ equation and VAR objects. If the Add-in is called from an equation object, it returns tables and vectors which contain cross-validation results for the forecasts of the base forms (e.g. non-transformed) of the dependent variable. If the Add-in is called from a VAR object, it returns cross-validation results for the forecast of the base forms of all the endogenous variables.
Cross-validation (or sometimes called forecast evaluation with a rolling technique) is a way of assessing the predictive performance by measuring the forecast errors. This Add-in provides 13 types of forecast error metrics including mean squared error (MSE) and mean absolute error (MAE).
The TSCVAL Add-in is a fantastic addition to the macro-economic tools available in EViews.
The TSVCAL Add-in, written by James Lamb and Rita Linets, performs rolling estimation and out-of-sample forecast evaluation from EViews’ equation and VAR objects. If the Add-in is called from an equation object, it returns tables and vectors which contain cross-validation results for the forecasts of the base forms (e.g. non-transformed) of the dependent variable. If the Add-in is called from a VAR object, it returns cross-validation results for the forecast of the base forms of all the endogenous variables.
Cross-validation (or sometimes called forecast evaluation with a rolling technique) is a way of assessing the predictive performance by measuring the forecast errors. This Add-in provides 13 types of forecast error metrics including mean squared error (MSE) and mean absolute error (MAE).
The TSCVAL Add-in is a fantastic addition to the macro-economic tools available in EViews.
Quarterly Prize
The EViews Development Team has decided that the TSCVAL Add-in contributed most significantly to the usage of EViews this quarter. This quarter’s $500 prize goes to James Lamb and Rita Linets, congratulations!
For more information on writing Add-ins, you can read the Add-in chapter of the online
help or visit the Add-in writer’s forum.
If you would like to submit an Add-in, need more information on the Quarterly Prize, or have quesitons about writing Add-in for EViews, please email support@eviews.com.
If you would like to submit an Add-in, need more information on the Quarterly Prize, or have quesitons about writing Add-in for EViews, please email support@eviews.com.
1 Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach”, Quarterly Journal of Economics 120.1: 38.7-422.
2 What are the effects of monetary policy on output? Results from an agnostic identification procedure," Journal of Monetary Economics, Elsevier, vol. 52(2), pp. 381-419.
3 Primiceri, G.E. (2005): ‘Time Varying Structural Vector Autoregressions and Monetary Policy’, Review of Economic Studies 72, pp. 821-852.
4 Sancetta A. 2010. Recursive forecast combination for dependent heterogeneous data. Econometric Theory 26: 598—631.
5 Bates JM, Granger CWJ. 1969. The Combination of Forecasts. Operations Research Quarterly 20: 451--468.}.
2 What are the effects of monetary policy on output? Results from an agnostic identification procedure," Journal of Monetary Economics, Elsevier, vol. 52(2), pp. 381-419.
3 Primiceri, G.E. (2005): ‘Time Varying Structural Vector Autoregressions and Monetary Policy’, Review of Economic Studies 72, pp. 821-852.
4 Sancetta A. 2010. Recursive forecast combination for dependent heterogeneous data. Econometric Theory 26: 598—631.
5 Bates JM, Granger CWJ. 1969. The Combination of Forecasts. Operations Research Quarterly 20: 451--468.}.
No comments:
Post a Comment