Bayesian approach

Oil price shocks are the major source of economic instability in oil exporting developing countries, including Iran. In this paper a Multi Sector Dynamic Stochastic General Equilibrium model, with emphasis on optimization of oil sector as a producing sector is designed. Furthermore, an optimizing import sector is introduced into the model by considering the price rigidity in imported goods as a source of inefficiency in a New Keynesian open economy. The impact of oil price shocks on the dynamics of the economic variables is considered during 1988:1-2011:1. For this purpose, the Bayesian approach is used to estimate the model. The impulse response functions show that immediately after an oil price shock, output increases in the oil sector, while in the non-oil sector the result is reverse. Furthermore, GDP, consumption and inflation increase, while the employment and real exchange rate decreases immediately and finally, all the variables converge to their steady state values. JEL Classification: C61, D50, E12, Q43

Economists and policymakers are interested in how government spending affects economic performance. The purpose of this study is to examine the effect of government capital expenditures by component on the GDP of Iran's provinces between 2011 and 2019 using a Bayesian approach. To do so, after estimating over 500,000 regressions and averaging the Bayesian model's coefficients, the five sub-categories of industry, judiciary, energy, health and information and communications technology were identified as the most effective sub-categories of government capital expenditures on provincial GDP. Then, a hierarchical Bayesian panel model was specified and estimated to determine the extent to which and how each of these subsections affects the provinces' GDP. Monte Carlo simulations using Markov chains indicate that while judicial and health expenditures reduced production during the study period, expenditures on industry, energy, information and communications technology increased production in Iran's provinces.

This paper intends to model the volatilities of returns of 20 different cryptocurrencies using daily data from 08/03/2018 to 09/20/2022. The multivariate factor stochastic volatility model (MFSV) within the framework of the nonlinear space-state approach is used. In this method, the cryptocurrency return volatility is decomposed into volatility rooted in latent factors and idiosyncratic volatility, and the time-varying pairwise correlation and dynamic covariance matrix are estimated in four sub-periods. The MFSV model’s results revealed that each sub-period contains a distinct number of latent factors, 2, 5, 4 and 2, which generally have a favorable impact on all cryptocurrency volatilities. The time-varying positive correlations between the return volatility of all cryptocurrencies are confirmed. Indeed, the strongest pairwise correlations belong to Ethereum, Litcoin, EOS, and VET in each sub-period, respectively. The DOGE, DOGE, Filecoin, and XRP, on the other hand, showedthe weakest correlations . As the pairwise correlations of cryptocurrency volatilities get strenger, especially during descending periods, it seems that the benefits of diversifying a crypto portfolio are getting less and less over time.