Bayesian Vector Autoregressive Approach Research Articles

Economic Policy Uncertainty and Equity Fund Flows to India: A Bayesian Approach

We compare the impacts of economic policy uncertainty (EPU) and global economic policy uncertainty (GEPU)-related shocks on equity fund flows (EFFs) to India using a Bayesian vector autoregression approach. We find that (a) Indian EPU and GEPU are strongly and negatively related to EFF; (b) EFF are more sensitive to GEPU relative to Indian EPU; (c) evidence of trend-chasing behaviour by fund managers in India; and (d) GEPU is an important factor for forecasting returns on the Bombay Stock Exchange. Taken together, our findings indicate that EPU is important to understanding equity allocation decisions and returns in India. JEL Codes: F21, F39, G11

Investigating the impulse responses of renewable energy in the context of China: A Bayesian VAR Approach

This study applies a Bayesian Vector Auto Regression (VAR) technique to investigate the impulse responses (IR) of renewable energy (RE) on different macroeconomic variables, i.e., green finance (GF), investment in energy with private participation (IEPP), gross domestic product growth rate (GDPG), and foreign direct investment (FDI), in the context of China by using data from the years 2001–2020. Data normality and stationarity are validated before checking the VAR stability condition. The IR function captures the direct and reverse responses of RE with the shift of stipulated variables. Subsequently, it reveals the causal relationship between RE and the considered variables. The results are: (i) RE and GF are strongly endogenous, whereas IEPP and GDPG are weakly endogenous, and FDI is least endogenous; (ii) the response of RE is positive to GF and negative to FDI, IEPP, and GDPG. The reverse responses are quite similar except for IEPP, and (ⅲ) shocks from GF, GDPG, and FDI lead to more variability in RE than RE itself. Ultimately, it indicates RE's share in total energy output increases with the boost of GF and decreases with the amplification of FDI, IEPP, and GDPG. Inversely, RE's contribution boosts with the short-term increase in IEPP.

Domestic Borrowing and the Nigerian Economy: A Bayesian VAR Analysis

ABSTRACT: The Nigerian government has borrowed for more than three decades to fund the growing fiscal deficit based on a weak tax revenue base. After the repayment of the country's foreign loans in 2006, the borrowing structure has been dominated by domestic loans. Between 2006 and 2020, domestic borrowing has grown by more than 800 per cent. This study examines the effect of domestic borrowing on Nigeria's macroeconomic variables from 1975 to 1980. The variables examined against the effect of domestic borrowing include economic growth, financial development (broad money supply and credit to private sector), external borrowing, interest rate, inflation, industrial production, private investment, trade openness and population growth rate. The study utilised the Bayesian Vector Autoregressive (BVAR) model. The BVAR approach gives more realistic estimations than other VAR models as it considers prior information steady states during analysis. The study also adopts the Cholesky impulse response function and variance decomposition approach in the investigation. The study found a long-run equilibrium relationship between domestic borrowing and other macroeconomic variables. Also, domestic borrowing positively and significantly affected economic growth, industrial production and inflation. However, domestic borrowing negatively and significantly affects trade openness. Although the effect of domestic borrowing on the financial development variables (broad money supply and credit to the private sector) is negative, it is insignificant. The Bayesian VAR impulse response simulation demonstrates that the fiscal policy shock from domestic borrowing impacts the macroeconomic variables in line with the estimated regression results. As a follow-up, the variance decomposition results show that the decomposition of the shocks from government domestic borrowing is borne more by external borrowing, broad money supply and interest rate in the order of magnitude. The study recommends a restriction in monetary policy to benchmark and closely checkmate the aggressive spending from domestic borrowing. Also, the Nigerian government should improve investment in infrastructure to enhance private sector productivity and expand taxes revenue base to mitigate the reliance on borrowing as a source of funding for fiscal deficit. Furthermore, the country's productivity must be coordinated by the government to expand the sources of foreign exchange earnings to enhance stability in the country's external sector.

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.

The effects of corruption on growth, human development and natural resources sector: empirical evidence from a Bayesian panel VAR for Latin American and Nordic countries

PurposeThe purpose of this paper is to analyze the effects of corruption on economic growth, human development and natural resources in Latin American and Nordic countries.Design/methodology/approachUsing the hierarchical prior of Gelman et al. (2003), a Bayesian panel Vector AutoRegression (VAR) model is estimated. In addition, two alternative approaches are considered, namely, a panel error correction VAR model and an asymmetric panel VAR model.FindingsThe results reveal some relevant contrasts: (1) in Latin America there is support for the sand the wheels hypothesis in Bolivia and Chile, support for the grease the wheels hypothesis in Colombia and no significant impact of corruption on growth in Brazil and Peru, while in Nordic countries the response of growth to shocks in corruption is negative in all cases; (2) corruption negatively affects human development in all countries from both regions; (3) corruption tends to spur natural resources sector in Latin American countries, while it is detrimental for natural resources sector in Nordic countries.Research limitations/implicationsThe panel VAR approach uses recursive scheme identification. The authors have analyzed robustness using alternative ordering of the variables. The authors also have followed two alternatives suggested by the Referee: a panel error correction VAR model and a panel asymmetric VAR model. However, another more sophisticated identification scheme could be used. Also other variables could be introduced in the VAR model.Practical implicationsRegardless of the issue of the “grease” vs the “sand the wheels” debate, corruption should be reduced because it is anyway harmful for human development. The differences in the results for Latin American and Nordic countries show that the effects of corruption have to be assessed considering the different institutional and economic conditions of the countries analyzed.Social implicationsGovernments should seek to reduce corruption because, despite corruption can have mixed effects on economic growth in some contexts, it is anyway harmful for human development. Besides, the finding that in some Latin American countries more activity in the extractive industries is generated by means of corruption confirm the association between corruption and extractivism found by Gudynas (2017) and can explain why there are issues of environmental damage and social conflict linked to natural resources in those countries.Originality/valueThe present study contributes to the literature by presenting evidence on the effects of corruption on growth, human development and natural resources sector in Latin American and Nordic countries. It is the first study on economics of corruption which directly compares Latin American and Nordic countries. This is relevant because there are important differences between both regions since Latin American countries tend to suffer from widespread corruption, while the Nordic ones have a high level of transparency. It is also the first in using a Bayesian panel VAR approach in order to evaluate the effects of corruption.

Globalisation, income inequality and poverty relationships: evidence from Mexico, Indonesia, Nigeria and Turkey

PurposeThe study examines the dynamic relationship among globalisation, income inequality and poverty in Mexico, Indonesia, Nigeria and Turkey (MINT countries) between 1980 and 2018.Design/methodology/approachA Bayesian vector autoregressive (BVAR) approach is used as a technique of estimation hanging on the fact that the method uses prior distribution for the estimated parameters.FindingsThe results show that globalisation is a strong predictor of itself in all the MINT countries only in the short run. In the long run, income inequality and poverty strongly influence globalisation, respectively, in Indonesia and Turkey, while globalisation still has more impact on itself in Nigeria. Income inequality has a strong endogenous impact on itself in Mexico and Indonesia over the time horizon, whereas globalisation and poverty are strong predictors of income inequality in the long run in Nigeria and Turkey, respectively. Also, poverty strongly influences itself in all the MINT countries in all the periods, meaning that poverty begets itself in all the MINT countries, except for Indonesia in the long run.Practical implicationsThe study suggests that all the MINT countries should ensure political stability and a strong institutional framework to gain from the process of globalisation and to experience reductions in the levels of income inequality and poverty.Originality/valueThis study is distinct from other studies in the sense that an overall globalisation index (GBI) as used by Dreher et al. (2008) is used for the globalisation variable, and the Multidimensional Poverty Index (MPI) is used to capture poverty in all the MINT countries. Also, the research paper uses a BVAR approach as against the classical VAR, and this helps in solving over-fitting problems.

Quantifying the relationship between seaborne trade and shipping freight rates: A Bayesian vector autoregressive approach

We employ a Bayesian Vector Autoregressive methodology, to counter the issue of data availability, and explore the relationship between seaborne commodity trade and freight rates. Our results show three important insights: first and foremost, the quantity of seaborne commodity trade has a strong impact on the Baltic Dry Index and the Baltic Dirty Tanker Index, but not on the Baltic Clean Tanker Index, most likely due to the fact that clean tankers can simultaneously operate both in the clean and the dirty sectors. Second, a shock in the price of brent oil has the expected positive response from the Baltic Dry Index, while its relationship with the Baltic Clean Tanker Index and the Baltic Dirty Tanker Index is negative as, in this case, tanker vessels can operate as floating storage units. Third, a relationship between the freight indices appears to hold as a change in one could spill over to the other.

The Transmission of Monetary Policy in South Africa Before and After the Global Financial Crisis

AbstractThis paper examines whether the transmission mechanism of monetary policy in South Africa has changed after the global financial crisis (GFC). We use a Bayesian vector autoregressive (BVAR) model with Minnesota priors and 15 monthly variables, extending the system of Christiano, Eichenbaum, with Evans (1999). The benefit of the BVAR approach is that it can accommodate a large cross section of variables without running out of degrees of freedom. To identify the change in the transmission process, we divide the sample size into two subsamples, namely the pre‐GFC period (March 2001 to August 2008) and the post‐GFC period (September 2008 to February 2016). The results indicate that a change in the transmission of monetary policy occurred after the GFC. The magnitude of the effect of a monetary policy shock on output is considerably greater in the pre‐GFC period compared to the post‐GFC period. Moreover, the impact of a policy shock on inflation is not statistically significant in the post‐GFC period. The variance decomposition shows that the interest‐rate channel has possibly weakened in the post‐GFC period.

A Bayesian vector autoregression-based data analytics approach to enable irregularly-spaced mixed-frequency traffic collision data imputation with missing values

Traffic collision data are always collected in irregularly spaced and mixed frequency. Conventional treatment on these kinds of data, for instance, aggregating the high-frequency data into the lower frequency, can lead to the loss of relevant information of high-frequency data, and introduce potential temporal instabilities. A novel Bayesian vector autoregression approach is proposed to address this problem. An unevenly-spaced traffic collision data with missing values, containing all collisions in different severities that occurred on the state highways in Washington State from January 2006 to December 2016, is selected in this study the impacts of transportation-, weather- and socioeconomic-related characteristics on traffic collisions. A Gibbs sampler is used to conduct Bayesian inference for model parameters and unobserved high-frequency variables. Results show that the model has a fairly superior fit accuracy, and is able to capture the unobserved heterogeneity in the dataset. The proposed VAR also demonstrates better performance than other missing value imputation techniques, including linear regression, predictive mean matching, k-nearest neighbors, and random forests. This study provides potential in the guidance of model construction that considers the mixed-time-series nature of data.

Shadow Banking and Canada’s Monetary Policy

We analyze the link between monetary policy and the growth of shadow banks, and by extension, financial stability in Canada. Using monthly financial market data from 1991-2015 in a two-stage time-varying coefficient Bayesian vector autoregression approach, we find that greater shares of shadow bank deposits and business loans reduce the effectiveness of monetary policy. Further investigations indicate that these drags are the result of deposits and credit shifting between shadow banks and traditional banks. We also find that contractionary monetary policy increases financial instability by shifting household mortgage loans and business loans from chartered banks to shadow banks.

Heterogeneous Impact of Monetary Policy on Regional Economic Activity: Empirical Evidence for Poland

ABSTRACT This article investigates the regional impact of monetary policy shocks in Poland. Regional responses to monetary policy shocks are estimated using Bayesian vector autoregressive approach, which allows us to model the entire panel of Polish regions through the imposition of a shrinkage prior. We then investigate factors underlying the asymmetric regional responses. We show that the regional differentiation in output response is stronger than those of unemployment or inflation. The asymmetry seems to be related to the regional industrial structure and demographic features.

A BVAR Approach to the Impact of Policy Uncertainty on International Trade and Investment

This article investigates the impact of economic policy uncertainty (EPU) on foreign trade and investment into India relative to other macroeconomic factors with reference to the Indian pharmaceutical industry. It employs the Bayesian vector autoregression (BVAR) approach to construct a model and then study the said impact by the impulse response function (IRF) analysis. In a Bayesian procedure, the parameters are treated as random variables and their posterior distribution is estimated via the imposition of prior beliefs on their distribution, which makes the analysis more robust when combined with vector autoregression. The IRF analysis shows no substantial protracted impact of policy uncertainty, inflation and interest rates on the pharma exports. Also, inflation seems to have a more pronounced impact on FDI as compared to the interest rate changes contemporaneously.

Measuring Brazilian Economic Uncertainty

We propose a measure of economic uncertainty, the Brazilian Economic Uncertainty Indicator, based on the news as well as business forecasts. The index expands the variety of newspapers handled by Baker et al. (Q J Econ 131(4):1593–1636, 2016) for Brazil. Our indicator captures Brazilian recent events such as the corruption scandals, the fiscal and economic crisis, the 2016 impeachment, the 2008 financial crisis and the 2002 presidential elections as moments of high uncertainty. An econometric study using a Bayesian Vector Autoregressive approach was carried out and revealed that uncertainty shocks cause an economic downturn in subsequent periods, as emphasized in the relevant literature.

Forecasting Household Credit in Kenya Using Bayesian Vector Autoregressive (BVAR) Model

This research paper use Bayesian VAR framework to forecast the household credit in the dynamic market of foreign remittances inflow to Kenya. The Bayesian VARs model in this study employs the sims-Zha prior to estimate. Bayesian vector autoregressive (BVAR) uses Bayesian methods to estimate a vector autoregressive (VAR). In that respect, the difference with standard VAR models lies in the fact that the model parameters are treated as random variables, and prior probabilities are assigned to them. This study employed data from the Kenyan Market for the period January 2005-December 2017. The forecast results were compared with the standard ARIMA model and the findings confirm that the BVAR approach outperforms the ARIMA model. Financial institutions can therefore use Bayesian VAR and other Bayesian models in predicting credit uptake given several micro-economic conditions. Banks should also find ways of tapping into these remittances especially those that pass through informal channels to improve their earnings from processing fees and also enhance the financial inclusion agenda through increasing account opening and loan uptake.

Comments on “Short-term inflation projections: A Bayesian vector autoregressive approach”

Comments on “Short-term inflation projections: A Bayesian vector autoregressive approach”

Short-term inflation projections: A Bayesian vector autoregressive approach

In this paper we construct a large Bayesian Vector Autoregressive model (BVAR) for the Euro area that captures the complex dynamic inter-relationships between the main components of the Harmonized Index of Consumer Prices (HICP) and their determinants. The model generates accurate conditional and unconditional forecasts in real-time. We find a significant pass-through effect of oil-price shocks on core inflation and a strong Phillips curve during the Great Recession.

Forecasting government bond yields with large Bayesian vector autoregressions

We propose a new approach to forecasting the term structure of interest rates, which allows to efficiently extract the information contained in a large panel of yields. In particular, we use a large Bayesian Vector Autoregression (BVAR) with an optimal amount of shrinkage towards univariate AR models. The optimal shrinkage is chosen by maximizing the Marginal Likelihood of the model. Focusing on the US, we provide an extensive study on the forecasting performance of the proposed model relative to most of the existing alternative specifications. While most of the existing evidence focuses on statistical measures of forecast accuracy, we also consider alternative measures based on trading schemes and portfolio allocation. We extensively check the robustness of our results, using different datasets and Monte Carlo simulations. We find that the proposed BVAR approach produces competitive forecasts, systematically more accurate than random walk forecasts, even though the gains are small.

A comparison of alternative rental forecasting models: empirical tests on the London office market

The study examines four alternative rental forecasting models in the context of the London office market. The forecasting ability of an ARIMA model, a Bayesian Vector Autoregression approach, an OLS based single equation model and a simultaneous equation model are compared and contrasted. The models are estimated using the CB Hillier Parker London Office index over the period 1977- 1996, with out-of-sample testing undertaken on the following three years of data. Diagnostic testing is also conducted on the alternative models. The findings reveal that the Bayesian VAR model produces the best forecasts, while the ARIMA model fails to pick up on the large uptake in rental values during the testing period.

Policy shocks and the Canadian macroeconomy: A Bayesian vector autoregression approach

In this paper we estimated a nine-variable Canadian monthly macroeconometric model within the context of a Bayesian vector autoregression (BVAR) innovation accounting technique over the period 1971:01 to 1988:12. Employing impulse response functions and variance decomposition measures we examined the monthly responses of Canadian key macroeconomic variables to unanticipated shocks in policy variables. Contrary to the results of the Dungan and Wilson study, which contradicts the ineffectiveness proposition of expenditure (or fiscal) policy of the standard new neoclassical model, our results support the ineffectiveness proposition of the latter model. The impact of monetary policy on real GDF in our study seemed to be quite strong and persistent.