Premium
Time varying structural VARs with sign restrictions: The case of Taiwan
Author(s) -
Chin KuoHsuan
Publication year - 2020
Publication title -
bulletin of economic research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.227
H-Index - 29
eISSN - 1467-8586
pISSN - 0307-3378
DOI - 10.1111/boer.12217
Subject(s) - vector autoregression , econometrics , volatility (finance) , economics , bayesian vector autoregression , stochastic volatility , sign (mathematics) , impulse response , structural vector autoregression , autoregressive model , bayesian probability , constant (computer programming) , inflation (cosmology) , mathematics , monetary policy , statistics , computer science , keynesian economics , mathematical analysis , physics , theoretical physics , programming language
I apply a Bayesian approach to a time‐varying structural vector autoregression model with stochastic volatility (TVP‐SVAR‐SV) to study the time‐varying nature of the Taiwanese economy. In particular, the structural parameters are identified via the sign information in a three‐variable VAR system. The estimated results show that TVP‐SVAR‐SV model has the best fit to the data, compared to the time‐varying parameters VAR model with constant volatility and a classical VAR model with constant parameters and volatilities. Moreover, I find the time‐varying contemporaneous relationship between the output growth and inflation rates, particularly significant before the year 2000. Lastly, the impulse responses and the volatilities of all the variables are found to be time‐varying.