Bayesian VAR Research Articles

Fiscal effects of exchange rate devaluation and capital flows to emerging countries

In the assessment of governments’ fiscal performance, exchange rates play some roles while capital movements could serve as determinant of fiscal discipline. This study examined the effects of exchange rate devaluation, and capital inflows, on budgetary spending, and the interactions among the variables using the Bayesian vector autoregression (BVAR) and sys-generalized method of moments (GMM) estimators with 1,184 panel observations. The study covers 37 emerging nations. The variables had a co-integrating connection, demonstrating a long-run link between the variables studied. Having executed the Gibbs sampling for simulation efficiently, our Markov Chain Monte Carlo (MCMC) simulation converged appropriately. The sampling efficiency parameter is equal to 0.96257, close to 1. The Monte Carlo standard errors (MCSE) are extremely low at 0.000 with an implication of adequate precision in the BVAR model estimation. The results disclose that a 1 percent devaluation shock compressed fiscal spending by 0.56 percent and a shock to capital inflows stimulated 0.99% growth in fiscal spending. The 95 percent credible interval suggests a considerable size of effects on devaluation and capital flows. Accordingly, managing the exchange rate can be a valuable tool for managing capital shortage in Africa. Rather than increase government spending, governments should concentrate on revenue generation by utilizing an effective exchange rate policy to influence the national pattern of product diversification.

Is the US Phillips curve stable? Evidence from Bayesian vector autoregressions*

AbstractIt has been claimed that the fall in US inflation during the Great Recession was surprisingly small. One possible explanation for this is that the Phillips curve is unstable and that its slope was lower around the Great Recession. We investigate the importance of time‐varying parameters using Bayesian vector autoregressions for inflation and unemployment. We find support for time variation in the inflation equation and an unstable Phillips curve that was somewhat flatter between 2005 and 2013. However, conditional forecasts mostly suggest that inflation was not unexpectedly high around the Great Recession, which puts the claim of a “missing disinflation” into question.

Unconventional Monetary Policy: Diversity of Measures and Information Effects

The paper presents an analysis of the consequences of the European Central Bank’s (ECB) unconventional monetary policy measures from 2007 to 2019. Econometric estimates of the effectiveness of the ECB’s policy can be biased for two reasons: first, the entire range of adopted measures may not be considered, and second, unexpected policy actions can prompt economic agents to revise their expectations and perceptions of the ECB’s monetary policy rule. Such an information shock might be mistakenly interpreted as a monetary shock. The approach proposed in the paper combines factor analysis and sign restrictions on high-frequency data in a Bayesian Vector Autoregression (BVAR) model. This approach yields findings that align with macroeconomic theory and offers a more comprehensive insight into the effects of different unconventional monetary policy measures on both financial and real variables. The identification of a monetary shock is carried out in parallel with the identification of an information shock. Unconventional monetary policy measures are approximated using high-frequency factors that influence specific segments of the yield curve. Unconventional ECB measures prove effective in stimulating business activity and price growth as well as in contributing to reduced financial stress. However, there is heterogeneity in their effects. The announcement of quantitative easing exhibits a small external lag with respect to the policy and leads to increased stock prices and lending volumes, indicating the effectiveness of the portfolio rebalancing and credit channels. For measures related to the shorter end of the yield curve, the external lag of the policy can extend up to a year. These measures strongly impact price growth due potentially to effective management of economic agents’ expectations. The consequences of information shocks are moderate and manifest prominently only in the initial months. This limits their ability to offset the effects of monetary policy and does not hinder the ECB’s goal of achieving inflation targets

Nowcasting and Forecasting Key Russian Macroeconomic Variables With the MFBVAR Model

This paper examines the quality of nowcasts and forecasts for Russian GDP and its components (in constant and current prices) using a mixed-frequency Bayesian vector autoregression model (MFBVAR) which is currently one of the most advanced time series forecasting models. It enables use of quarterly and monthly frequency data within a single monthly frequency VAR model in a statespace form while taking into account the intra-quarter dynamics of monthly indicators; this approach improves forecasting accuracy when new monthly data is published. The MFBVAR model’s resistance to the jagged edge problem is especially important for real-time forecasting, and it can incorporate a large number of predictors because of its Bayesian estimation with a Minnesota-type prior distribution. The paper sets up three experiments with differing availability of monthly data in order to test pseudo out-of-sample nowcasting and forecasting. The MFBVAR model exhibits statistically significant outperformance compared to a naïve benchmark, as well as to ARIMA and quarterly BVAR models, in nowcasting and forecasting a few steps ahead for GDP, consumption and foreign trade variables. The test sample is also quite representative and covers two crisis periods, specifically 2015 and 2020. In both crises, the model accurately estimates the scale of the recession and recovery of economic activity. Nevertheless, there was no significant improvement in the quality of forecasts when new available monthly data was introduced.

Uncertainty and demand for business loans: A study of selected countries in the euro area

This paper studies the effect of uncertainty shocks on the demand for business loans in individual euro area countries. The results of Bayesian vector autoregression (VAR) model impulse response functions show that in some countries the overall demand for business loans, and particularly the demand for business loans for fixed-investment financing, respond significantly negatively to the shock.

Economy Using a Large Non-linear Bayesian VAR Model

Economy Using a Large Non-linear Bayesian VAR Model

Conditional Forecasts in Large Bayesian VARs with Multiple Soft and Hard Constraints

Conditional Forecasts in Large Bayesian VARs with Multiple Soft and Hard Constraints

Forecasting the democratic republic of the Congo macroeconomic data with the Bayesian vector autoregressive models

The purpose of this paper is to emphasize the role of the Bayesian Vector Autoregressive models (VAR) in macroeconomic analysis and forecasting. To help the policy-makers to do better, the Bayesian VAR models are considered more robust and valuable because they put in the model the mathematician’s beliefs or priors and the data. By using BVAR(1), we get the main results: (i)the best out sample point forecasts; (ii) the exchange rate shock contributes more to inflation; (iii) the inflation shock has high effects on exchange rate innovation. These results are due to the dollarization of this small open economy.

Estimating long-term and short-term impact of COVID-19 activity restriction on regional highway traffic demand: A case study in Zhejiang Province, China

Since the outbreak of COVID-19 in China in late 2019, government administrators have implemented traffic restriction policies to prevent the spread of COVID-19. However, highway traffic volumes obtained from ETC data in some provinces did not return to the levels of previous years after the end of the traffic restriction policy, suggesting that traffic restriction policy may have long-term effects. This paper proposed a method that analyzes traffic restriction policies' long-term and short-term impact on highway traffic volume under COVID-19. This method first analyzes the long-term and short-term impacts of traffic restriction policies on the highway traffic volume using the Prophet model combined with the concept of traffic volume loss. It further investigates the relationship between COVID-19 cases and the long-term and short-term impacts of the traffic restriction policy using Granger causality and the impulse response function of the Bayesian vector autoregressive (BVAR) model. The results showed that during the COVID-19 pandemic, highway traffic in Zhejiang Province decreased by about 95.5%, and the short-term impact of COVID-19 cases was most pronounced on the second day. However, the long-term effects were relatively small when the traffic restriction policy ended and was verified by data from other provinces. These results will provide decision support for traffic management and provide recommendations for future traffic impact assessments in the event of similar epidemics.

The dynamic complementarity of renewable energy sources: A Bayesian vector autoregressive approach

ABSTRACT The climate factors are the main driving forces of variability of renewable energy sources. In facing the safety challenges of the large-scale grid integration of renewable energy, it is important to understand the effects of climate shocks on the generation of the renewables as well as if the complementarity among different types of renewable energy exist and could be employed to smooth out the production and better match with the load. We employ a Bayesian vector autoregressive (BVAR) model along with impulse response function and forecast error variance decomposition to investigate the dynamic complementarity among renewable energy in the face of different climate shocks. The proposed methodology is applied over Qinghai employing the power generation data as well as meteorological data. The results show that there is greater general complementarity between solar power and hydropower. Wind power and solar power, and wind power and hydropower, exhibit dynamic complementarity when there are fluctuations in precipitation and wind speed, respectively. China should make efficient use of the natural complementarity advantages between renewable energy sources, carry out joint optimal dispatching, and accelerate the construction of a clean, low-carbon, safe and efficient energy system.

Bayesian VARs and prior calibration in times of COVID-19

Abstract This paper investigates the ability of several generalized Bayesian vector autoregressions to cope with the extreme COVID-19 observations and discusses their impact on prior calibration for inference and forecasting purposes. It shows that the preferred model interprets the pandemic episode as a rare event rather than a persistent increase in macroeconomic volatility. For forecasting, the choice among outlier-robust error structures is less important, however, when a large cross-section of information is used. Besides the error structure, this paper shows that the standard Minnesota prior calibration is an important source of changing macroeconomic transmission channels during the pandemic, altering the predictability of real and nominal variables. To alleviate this sensitivity, an outlier-robust prior calibration is proposed.

Density forecasting with Bayesian Vector Autoregressive models under macroeconomic data uncertainty

SummaryMacroeconomic data are subject to data revisions. Yet, the usual way of generating real‐time density forecasts from Bayesian Vector Autoregressive (BVAR) models makes no allowance for data uncertainty from future data revisions. We develop methods of allowing for data uncertainty when forecasting with BVAR models with stochastic volatility. First, the BVAR forecasting model is estimated on real‐time vintages. Second, the BVAR model is jointly estimated with a model of data revisions such that forecasts are conditioned on estimates of the ‘true’ values. We find that this second method generally improves upon conventional practice for density forecasting, especially for the United States.

Comparing stochastic volatility specifications for large Bayesian VARs

Large Bayesian vector autoregressions with various forms of stochastic volatility have become increasingly popular in empirical macroeconomics. One main difficulty for practitioners is to choose the most suitable stochastic volatility specification for their particular application. We develop Bayesian model comparison methods–based on marginal likelihood estimators that combine conditional Monte Carlo and adaptive importance sampling–to choose among a variety of stochastic volatility specifications. The proposed methods can also be used to select an appropriate shrinkage prior on the VAR coefficients, which is a critical component for avoiding over-fitting in high-dimensional settings. Using US quarterly data of different dimensions, we find that both the Cholesky stochastic volatility and factor stochastic volatility outperform the common stochastic volatility specification. Their superior performance, however, can mostly be attributed to the more flexible priors that accommodate cross-variable shrinkage.

The European Central Bank and green finance: How would the green quantitative easing affect the investors' behavior during times of crisis?

In July 2021, the European central bank (ECB) announced the application of new environmental criteria to purchase private assets as part of its Quantitative Easing (QE) program. Using a Bayesian VAR model with time varying parameters and stochastic volatility (TVP-BVAR-SV), we investigate the transmission of Green bond shocks to the stock market during the pre-and-post COVID-19 pandemic. We document a nonlinear relation between the green bonds and the green equities. Our findings suggest that the ECB's Green QE can drive investors towards green investment in the stock market through the green bond market during the non-crisis period. However, we show that the proper transmission of Green QE shocks to the stock market depends on the economic conditions and could not be effective during the crisis period. Our results also support previous findings that state the growing demand for sustainable investing after COVID-19. These findings have important implications for investment professionals, policymakers, and environmentally concerned actors.

Supply shocks and monetary policy responses in emerging economies

Supply shocks bring about important dilemmas for monetary policy in emerging economies. We compute monetary policy responses to supply shocks using quarterly data and a Bayesian panel VAR for 24 emerging economies during the period 2004–2019. In this framework, we identify supply shocks as unexpected and temporary total factor productivity innovations which are orthogonal to demand related shocks. We highlight three results from this econometric exercise. First, monetary policy is, in average, procyclical in emerging economies after temporary supply shocks. Second, monetary policy is more procyclical in fixed than in flexible exchange-rate regimes. Third, monetary policy is more procyclical in economies with a higher degree of financial openness. The latter result is a central dilemma for emerging economies during supply shocks, since it implies that financial openness prevents less procyclical monetary policy reactions due to the trade-off between exchange rate and income volatility.

INDONESIA’S FINANCIAL STRESS EVENTS AND MACROECONOMIC DYNAMICS

In this study, we use a Markov-Switching Bayesian Vector AutoRegression model to investigate the episodic relationship between financial stress and the key macroeconomic variables in the case of Indonesia. We find different nature of relationships among Indonesia’s real sector variables (household consumption expenditure and consumer price index), financial sector variables (interbank money market rate) and the policy variable (broad money supply during the times of high and low financial stress). Regime changes occurred on several occasions, including during the 2008 global financial crisis period and at the beginning of the COVID-19 pandemic.

INDONESIA’S FINANCIAL STRESS EVENTS AND MACROECONOMIC DYNAMICS

In this study, we use a Markov-Switching Bayesian Vector AutoRegression model to investigate the episodic relationship between financial stress and the key macroeconomic variables in the case of Indonesia. We find different nature of relationships among Indonesia’s real sector variables (household consumption expenditure and consumer price index), financial sector variables (interbank money market rate) and the policy variable (broad money supply during the times of high and low financial stress). Regime changes occurred on several occasions, including during the 2008 global financial crisis period and at the beginning of the COVID-19 pandemic.

Large Time‐Varying Volatility Models for Hourly Electricity Prices*

AbstractWe study the importance of time‐varying volatility in modelling hourly electricity prices when fundamental drivers are included in the estimation. This allows us to contribute to the literature of large Bayesian VARs by using well‐known time series models in a large dimension for the matrix of coefficients. Based on novel Bayesian techniques, we exploit the importance of both Gaussian and non‐Gaussian error terms in stochastic volatility. We find that using regressors as fuel prices, forecasted demand and forecasted renewable energy is essential to properly capture the volatility of these prices. Moreover, we show that the time‐varying volatility models outperform the constant volatility models in both the in‐sample model‐fit and the out‐of‐sample forecasting performance.

Effects of macro-financial shocks on bank liquidity creation: evidence from Mongolia

This paper examines the effects of macro-financial shocks on bank liquidity creation in Mongolia, a developing and commodity-exporting economy, using a structural Bayesian vector autoregression (SBVAR). Ten structural shocks (three external and seven domestic shocks) are identified using a triangular factorization. The main results are (i) the measured bank liquidity creation (as a share of total assets) is procyclical and negatively correlated with the liquidity coverage ratio in the banking sector; (ii) external shocks (the US monetary policy, Chinese economic activity, and changes in the global commodity market) have statistically significant and economically major effects on the liquidity creation. Lending rate, NPL ratio, foreign exchange reserves, and competition in the banking sector are key domestic determinants of liquidity creation; (iii) liquidity creation’s main components respond differently to the US federal funds rate and lending rate shocks. External and financial shocks have contributed to the liquidity creation passing through movements in (illiquid assets + liquid liabilities) component; (iv) monetary policy can be an effective counter-cyclical policy instrument in stabilizing the economic and financial cycles.

Household deposits and consumer sentiment expectations: Evidence from Eurozone

This paper examines whether new refined measures of consumer sentiment expectations exert a significant effect on household deposit flows across Eurozone’s member countries. The new measures of sentiment-based expectations of consumers are constructed following a principal component analysis of selected individual answers as part of Eurostat’s consumer sentiment survey across the Eurozone’s member countries. Employing a Bayesian panel VAR model to increase the accuracy of our results, we document that a (positive) shock on consumers’ sentiment expectations triggers an imminent positive response of household deposits that gradually reverses during the next 2–3 months. This main result extends and complements traditional explanations for depositors’ behavior in the extant literature. Our results have important implications for Eurozone’s households, policymakers, central bankers, and commercial banks.