FACTOR-AUGMENTED VAR MODEL FOR IMPULSE RESPONSE ANALYSIS

Structural Vector Autoregressive Models are largely used in the economic literature in order to uncover the way the transmission of a shock in one macroeconomic variable is seen in the evolution of other macroeconomic variables. These models have the advantages that a small number of restrictions needs to be set in order to be able to estimate the transmission. Despite this, for small open economies the estimated results might be affected by the so called "puzzles" or the initial data set may omit important information which is a major disadvantage for studies on the effect of monetary policy innovations. The above mentioned issues are solved by Factor Augmented Vector Autoregression models, therefore the purpose of this paper is to estimate this type of model on a sample of 20 variables for the Romanian economy, by using quarterly data over the period: 2000:Q1 and 2014:Q2. The factors are estimated using Principal Component Analysis and Bayesian inference is used as methodology in order to compute the parameters of the model.