scholarly journals On the Performances of Classical VAR and Sims-Zha Bayesian VAR Models in the Presence of Collinearity and Autocorrelated Error Terms

2016 ◽  
Vol 06 (01) ◽  
pp. 96-132
Author(s):  
M. O. Adenomon ◽  
V. A. Michael ◽  
O. P. Evans
Keyword(s):  
Var Models ◽  
Bayesian Var ◽  
Autocorrelated Error ◽  
Error Terms

scholarly journals Forecasting Performances of the Reduced Form VAR and Sims-Zha Bayesian VAR Models when the Multiple Time Series are Jointly Influenced by Collinearity and Autocorrelated Error

2018 ◽  
Author(s):  
Monday Osagie Adenomon ◽  
Benjamin Agboola Oyejola
Keyword(s):  
Time Series ◽  
Data Structure ◽  
Multiple Time ◽  
Multiple Time Series ◽  
Var Models ◽  
Short Term ◽  
Series Length ◽  
Bayesian Var ◽  
Autocorrelated Error ◽  
Different Levels

The goal of VAR or BVAR is the characterization of the dynamics and endogenous relationships among time series. Also the VAR models are known for their applications to forecasting and policy analysis. This paper compare the performance of VAR and Sims-Zha Bayesian VAR models when the multiple time series are jointly influenced by different levels of collinearity and autocorrelation in the short term (T=16, 32, 64 and 128). Five levels (-0.9,-0.5, 0,+0.5,+0.9) of collinearity and autocorrelation were considered and the results from the simulation study revealed that VAR(2) model dominated for no and moderate levels of autocorrelation (-0.5, 0, +0.5) irrespective of the collinearity level except in few cases when T=16. While the BVAR models dominated for high autocorrelation levels (-0.9 and +0.9) irrespective of the collinearity level except in few cases when T=128. The performance of the models varies at different levels of the collinearity and autocorrelated error, and also varies with the short term periods. Furthermore, the values of the RMSE and MAE criteria decrease as a result of increase in the time series length. In conclusion, the performance of the forecasting models depend on the time series data structure and the time series length. It is therefore recommended that the data structure and series length should be considered in using an appropriate model for forecasting.

Forecasting Performance Comparison by Using Power Transformation between VAR and Bayesian VAR Models

2014 ◽  
Vol 529 ◽  
pp. 621-624
Author(s):  
Syang Ke Kung ◽  
Chi Hsiu Wang
Keyword(s):  
Time Series ◽  
Industrial Production ◽  
Multivariate Time Series ◽  
Performance Comparison ◽  
Power Transformation ◽  
Var Models ◽  
Bayesian Var ◽  
Forecasting Performance ◽  
The U.S ◽  
Forecasting Practice

This article is devoted to examine the performance of power transformation in VAR and Bayesian VAR (BVAR) forecasts, in comparison with log-transformation. The effect of power transformation in multivariate time series model forecasts is still untouched in the literature. We examined the U.S. macroeconomic data from 1960 to 1987 and the Taiwan’s technology industrial production from 1990 to 2000. Our results showed that the power transformation provides outperforming forecasts in both VAR and BVAR models. Moreover, the non-informative prior BAVR with power transformation is the best predictive model and is recommendable to forecasting practice.

scholarly journals A Comparison of Different Short-Term Macroeconomic Forecasting Models: Evidence from Armenia

2016 ◽  
Vol 5 (2) ◽  
pp. 81-99 ◽  
Author(s):  
Karen Poghosyan
Keyword(s):  
Mean Squared Error ◽  
Small Scale ◽  
Var Models ◽  
Short Term ◽  
Macroeconomic Forecasting ◽  
Bayesian Var ◽  
Out Of Sample ◽  
Competing Models ◽  
Bayesian Factor ◽  
Short Term Forecasting

Abstract We evaluate the forecasting performance of four competing models for short-term macroeconomic forecasting: the traditional VAR, small scale Bayesian VAR, Factor Augmented VAR and Bayesian Factor Augmented VAR models. Using Armenian quarterly actual macroeconomic time series from 1996Q1 – 2014Q4, we estimate parameters of four competing models. Based on the out-of-sample recursive forecast evaluations and using root mean squared error (RMSE) criterion we conclude that small scale Bayesian VAR and Bayesian Factor Augmented VAR models are more suitable for short-term forecasting than traditional unrestricted VAR model.

Forecasting hotel demand uncertainty using time series Bayesian VAR models

2018 ◽  
Vol 25 (5) ◽  
pp. 734-756 ◽  
Author(s):  
Apostolos Ampountolas
Keyword(s):  
Demand Uncertainty ◽  
Error Rates ◽  
Percentage Error ◽  
Revenue Maximization ◽  
Var Models ◽  
Forecasting Accuracy ◽  
Bayesian Var ◽  
Vector Autoregressive ◽  
Forecasting Performance ◽  
The Mean

Demand uncertainty is a fundamental characteristic of the hospitality industry. Hotel room inventory is fixed, and devising an accurate daily demand measurement is a key operational challenge. In practice, it is difficult to predict the industry stability and capture demand uncertainty, so the industry relies on demand estimates. This process of estimation affects revenue maximization, as it is sensitive to incremental costs. In this article, we implemented vector autoregressive (VAR) models and compared them to the Bayesian VAR to examine the accuracy of predicting demand. We evaluated the results using a new measure of forecasting accuracy, the mean arctangent absolute percentage error (MAAPE). The results generated from the forecasts confirm the significant improvement in forecasting performance that can be obtained using the Bayesian model. It is noteworthy that the VAR performs the best for the lower horizons. The results also suggest that MAAPE outperforms other existing accuracy measures, in terms of error rates.

Sign in / Sign up

Export Citation Format

Share Document