Bayesian modelling of real GDP rate in Romania
Journal Title: Computational Methods in Social Sciences - Year 2015, Vol 3, Issue 1
Abstract
Themain objective of this study is to model and predict the real GDP rate using Bayesian approach. A Bayesian VAR (BVAR), a Bayesian linear model and switching regime Bayesian models were employed for the real GDP rate, inflation rate and interest rate.From the set of variables that were connected to real GDP, for identifying the most relevant ones using the data for Romanian economy, we applied the selection algorithm based on stochastic search. Weight of revenues in GDP, weight of budgetary deficit in GDP, investment rate and inflation rate are the most correlated variables with the real GDP rate. The averages of posterior coefficients of models were used to make forecasts. For Romania on the horizon 2011-2014, the unrestricted switching regime models generated the most accurate forecasts.
Authors and Affiliations
Mihaela SIMIONESCU
Keywords
Related Articles
- EP ID EP101195
- DOI -
- Views 126
- Downloads 0
Foreign direct investments in Romania in EU28 framework
Foreign direct investments flows are perceived by economic policymakers and by economic researchers as one of the key-determinants of the process of adjustment and structural modernization of em...
Individual contributions to portfolio risk: risk decomposition for the BET - FI index
The paper applies Euler formula for decomposing the standard deviation and the Expected Shortfall for the BET - FI equity index. Risk attribution allows the decomposition of the total risk of t...
Using the median and the mean of the income to establish the poverty lines
One of the methods to estimate the poverty level inside a given population is based on how to define the poverty line values. Each person having his income under the poverty threshold will be considered to be poor. In th...
Developing a High Performance Software Library with MPI and CUDA for Matrix Computations
Nowadays, the paradigm of parallel computing is changing. CUDA is now a popular programming model for general purpose computations on GPUs and a great number of applications were ported to CUDA obtaining speedups of orde...
The issue of statistical power for overall model fit in evaluating structural equation models
Statistical power is an important concept for psychological research. However, examining the power of a structural equation model (SEM) is rare in practice. This article provides an accessible review of th...