Unrestricted VAR Research Articles

The Effects of Household Debt and Oil Price Shocks on Economic Growth in the Shadow of the Pandemic

In a sample of 34 countries during 1965Q2 to 2021Q3, this paper offers an empirical analysis of how household debt and oil price shocks influence economic growth in the shadow of the pandemic. We exploit the quarter lags inherent in the response of debt and the oil price to output to pin down the relationship between household debt, the oil price, and economic growth in an unrestricted panel VAR model. We find that household debt has a short-term positive impact on economic growth, and this impact is lagged, while oil price shocks have a negative effect on economic growth. Pandemic uncertainty has an obvious and positive effect on household debt, while it has an obvious and negative effect on economic growth and oil price. The results hold under several robustness tests.

Role of 2008 financial contagion in effecting the mediating role of stock market indices between the exchange rates and oil prices: Application of the unrestricted VAR

Role of 2008 financial contagion in effecting the mediating role of stock market indices between the exchange rates and oil prices: Application of the unrestricted VAR

Impact of remittances on economic growth in Republic of Kosovo

In the following analysis, a much debated topic has been analyzed with regard to remmitances and economic growth in Kosovo. The study is based on time series, quarterly data over the period 2010-2021. The methodology applied in the study is based on utilization of unrestricted VAR model and Granger Causality/Block exogeniety Wald test, in order to explore the direction of the causality between variables. Results show that there is bi-directional causality between remittances and economic growth in Kosovo in the short-run.

Modelling the electricity generation dynamics of Ghana. A Structural VAR Approach

This study estimates endogenous parameters to ascertain the dynamics in the electricity generation sector in Ghana. An unrestricted VAR model is utilized to examine the empirical ramifications of the Ghanaian electricity energy sector and power consumption. The data period ranges between 2002 02021. Precisely, the results detected structural long run and short run headwinds to the unrestricted models. The findings depict that the reactions of GDP growth rate and electricity from fossil fuels react directly to headwinds based on their correlation. Also, there is feedback relationship between GDP growth rate rand electricity from fossil fuel sources since they are significant, from the granger causality analysis. Furthermore, the impulse function depicts GDP growth rate reacts to exogenous headwinds and the impacts are enduring to an extended period. Furthermore, the Variance decomposition results backed this analysis via having renewable energy without hydropower explaining less than 1% variance due to shocks and total global greenhouse emissions further explains about 85% variance due to headwinds in the time period. Electricity from fossil fuels sources explains more than 100% of the variance owing to headwinds hitting the system. Implying the nation’s overreliance on conventional energy source. Installed renewable energy will grow to over 2500MW by 2036, growing at 57.8% %. Ultimately, this calls for a rapid transformation of energy landscape in the country to a sustainable pathway. This will inform policy discourses regarding diversifying Ghana’s energy mix. Ghana VAR Electricity Energy Conventional energy RES

Electricity, Exergy and Economic Growth in Mozambique

Electricity consumption and economic growth attracted the attention of researchers in several countries. Mozambique have few studies on this matter and non-related to useful exergy. This paper analyses the relationship between final electricity consumption, useful exergy and economic growth in Mozambique, from 1971 to 2014. The final electricity consumption data are expressed in GWh, from International Energy Agency database, which for Mozambique (until the date of writing) cover the period from 1971 to 2014. For this analysis an unrestricted VAR models were estimated to capture different types of effects, which are particularly important for an economy that underwent diverse phases and restructuring processes. Results indicate that there was a huge increase in final electricity consumption, from 2000 and 2001. Modelling with final electricity consumption and with useful energy illustrated that economic growth in Mozambique has not been influenced by final electricity consumption and useful exergy. However, economic growth induces final electricity consumption and useful exergy growth. The results suggests that sectors that induce economic growth in Mozambique are the least electricity intensive ones. The Police implications is that electricity access of those sectors that induce economic growth should be enhanced to keep them more competitive, and to sustain the growth.

Do Energy Prices drive Inflation in South Africa? An unrestricted VAR Approach

This study was aimed at investigating the effects of energy prices on inflation in South Africa. This was achieved by means of econometric analysis and annual time series data spanning from 1994 to 2020. The findings revealed that electricity tariffs and petrol prices exhibit a positive effect on inflation in South Africa. On the contrary, interest rates and the exchange rate as control variables were found to have a negative effect on the inflation rate. The impulse response function indicated that inflation responds positively to innovations in petrol prices, electricity tariffs and money supply in the medium and long term but responds negatively to innovations in interest rates and exchange rates. Further to this, the variance decomposition function revealed that variations in inflation are largely explained by its own innovations and partially explained by innovations in petrol prices, money supply and interest rates. Lastly, the granger causality analysis showed no evidence of causality between inflation and explanatory variables during the specified period. Given that energy prices place upward pressure on inflation, the study recommends a review of the current fuel and electricity tariff structure to provide the much-needed relief to households and businesses.

Impact of Exchange Rate on Gold Price in India

In this article we investigate the impact of rupee-dollar exchange rates on growth rate of gold price in India over the period 1970-2016. The investigation is done using some steps. In the first step we use Augmented Dickey-Fuller (ADF) and Phillip-Perron (PP) test to detect that the series is stationary or not. To show the long run association of the variables we apply Johansen Co-integration test. As there is no cointegration we apply unrestricted VAR method. Since the study variables are not cointegrated, there is no long run causality, there is only short run causality running from independent variables to dependent variables. Further the Granger Causality test determines whether growth rate of gold price depends on the growth rate of rupee dollar exchange rate or not. At last we also test the stability of the model using Eigen value statistic method. The main finding of this paper is that growth rate of gold price depends negatively and significantly on growth of rupee-dollar exchange rate, means that as rupee dollar exchange rate rises gold price decreases and vice versa. The study also shows that growth rate of gold price does not Granger cause to rupee-dollar exchange rate where as rupee- dollar exchange rate Granger cause to gold price in Indian economy. That is we can stabilise the Gold Price movement by controlling the exchange rate fluctuations.

Combining a DSGE Model with Variational Bayesian Neural Networks

In the spirit of the DSGE-VAR approach, I employed Temporal Difference Variational Auto-Encoder (TDVAE) proposed by Gregor et al. (2019) as a reduced timeseries model on which theoretical restrictions from a DSGE model are imposed. TDVAE is virtually a general-form stochastic state-space model implemented via variational Bayesian inference (VI) with neural networks. With the flexibility of neural networks, this DSGE-TDVAE approach tries to address the Consolo et al. (2009)’s criticism on DSGE-VAR that an unrestricted VAR may not represent the data correctly due to its shortage in expressiveness. The empirical results showed that forecast performance is certainly gained by combining the standard New Keynesian DSGE with TDVAE for the actual macroeconomic data of Japan. It confirmed that some support for the DSGE model is still found in the data even when evaluated against the more general statistical benchmark of TDVAE rather than VAR. At the same time, this indicated that the DSGE holds considerable degree of misspecification as the optimal intensity of DSGE restriction was small. Furthermore, DSGE-TDVAE achieved the superior forecasting performance to DSGE-VAR.

Investigating the relationship between CO2 emission and industrial production of bangladesh through the unrestricted vector auto regression methods

Industrial production units discharge large amounts of CO2 as manufacturing facilities directly use fossil fuels and more electricity than any other sector. Although the per capita emissions in the industrialized countries are typically as much as ten times the average in the developing countries, this study is aimed at analyzing the long-run association between CO2 emissions and industrial production in Bangladesh using an Unrestricted Vector Auto Regression approach. For this purpose, the study uses secondary data for the periods of 1960 to 2016 from world development indicators. The variables of interest are co2 emission and industrial production. In general CO2 data are measured in metric tons per capita and the industrial production index is used as the proxy of industrial production. Other econometric techniques, such as unit root test-ADF, Johanson Co-integration test, and OLS techniques are also applied. Firstly, a descriptive analysis finds that there has been a rapid fall in industrial output and co2 emission in 1971 which can be denoted as an adverse effect of the Independence war of Bangladesh. Despite that, industrial production and co2 emission are intended to increase at a positive slope till 2016. But the increasing rate of industrial production is significantly higher than the increasing rate of co2 emission in Bangladesh. Secondly, the Johanson Co-integration test results reveal that there is no long-run relationship between industrial production and CO2 emission in Bangladesh. But the results from Unrestricted VAR and Ordinary Least Square estimation confirm that CO2 emission one period lag has a negative and significant impact on the industrial production of Bangladesh, where the value of the coefficient is -16.01059. This means that if in the last year, co2 emission increased by 1 metric ton per capita, industrial production will be decreased by 16% in the current period. The study concludes that Bangladesh is running conscious industrial production considering energy conservation policies.

Education enrolment rate vs employment rate: Implications for sustainable human capital development in Nigeria

PurposeThe study examines the dynamic interrelationships among the school enrolment rates and the rate of employment (via unemployment rates) in Nigeria. Design/methodology/approachThe study employed Autoregressive estimates and an unrestricted VAR approach to analyze these relationships. FindingsThe study lends credence to the new-growth theory (i.e. endogenous models) that more investments in human capital, through education especially at higher levels, will allow human capital to evolve dynamically to increase long-run growth positively in Nigeria. This tendency engenders multiplier effects in stimulating sustainable human capital development given that education-driven growth facilitates employment in the short-run. However, to sufficiently sustain human capital development to generate employment in the long-run, it appears there is need to combine education with other cooperant factors such as social safety nets, good governance, private sector development and efficient utilization of human and physical resources. Originality/valueThe growth literature has been definitive on the role of human capital in achieving long-run economic growth. Therefore, investments in education have been identified as a vital channel for building human capital and achieving long run development objectives. Thus, in the nascent quest for sustainable development, this study takes the new growth theory a step higher by examining the modulating effects of education-driven growth (i.e. via school enrolments rates) in setting the pace for employment patterns in Nigeria.

Stationary Bootstrap Based Multi-Step Forecasts for Unrestricted VAR Models

This paper proposes a new asymptotically valid stationary bootstrap procedure to obtain multivariate forecast densities in unrestricted vector autoregressive models. The proposed method is not based on either backward or forward representations, so it can be used for both Gaussian and non-Gaussian models. Also, it is computationally more efficient compared to the available resampling methods. The finite sample performance of the proposed method is illustrated by extensive Monte Carlo studies as well as a real-data example. Our records reveal that the proposed method is a good competitor or even better than the existing methods based on backward and/or forward representations.

The impact of direct and indirect taxes on economic growth: an analytical approach from the Republic of Kosovo

The objective of this paper is to investigate the effects of direct and indirect taxes on economic growth of Kosovo, as a relatively newly-independent country with a transitional economy. The study covers a ten-year period with a monthly time series from 2006m1 to 2016m9. The time series have been tested for unit root through the employment of augmented Dickey-Fuller test, to determine if time series data are stationary or possess unit root. Through the utilisation of Johansen test for cointegration, the study analyses if models contain at least one cointegrating vector, thus indicating significant long-term relationship among direct, indirect taxes and economic growth in the country. Finally, to find the impact of direct and indirect taxes on economic growth, vector autoregressive model - unrestricted VAR has been employed. Results suggest a positive and significant effect of indirect taxes on real GDP in Kosovo.

Education Enrollment Rate vs Employment Rate: Implications for Sustainable Human Capital Development in Nigeria

The study examines the dynamic interrelationships among the school enrolment rates and the rate of employment (via unemployment rates) in Nigeria. The study employed Autoregressive estimates and an unrestricted VAR approach to analyze these relationships. The study lends credence to the new-growth theory (i.e. endogenous models) that more investments in human capital, through education especially at higher levels, will allow human capital to evolve dynamically and increase long-run growth in Nigeria. This tendency engenders multiplier effects in stimulating sustainable development given that education-driven growth facilitates employment. The growth literature has been definitive on the role of human capital in achieving long-run economic growth. Therefore, investments in education have been identified as a vital channel for building human capital and achieving long run development objectives. Thus, in the nascent quest for sustainable development, this study takes the new growth theory a step higher by examining the modulating effects of educational-driven growth (i.e. via school enrolments rates) in setting the pace for employment patterns in Nigeria.

The impact of direct and indirect taxes on economic growth: an analytical approach from the Republic of Kosovo

The objective of this paper is to investigate the effects of direct and indirect taxes on economic growth of Kosovo, as a relatively newly-independent country with a transitional economy. The study covers a ten-year period with a monthly time series from 2006m1 to 2016m9. The time series have been tested for unit root through the employment of augmented Dickey-Fuller test, to determine if time series data are stationary or possess unit root. Through the utilisation of Johansen test for cointegration, the study analyses if models contain at least one cointegrating vector, thus indicating significant long-term relationship among direct, indirect taxes and economic growth in the country. Finally, to find the impact of direct and indirect taxes on economic growth, vector autoregressive model - unrestricted VAR has been employed. Results suggest a positive and significant effect of indirect taxes on real GDP in Kosovo.

Affordable Housing in the Wake of Global Pandemics: A Reality or a Mirage the Kenyan Perspective?

Aim: This was an investigative study on affordable housing in the wake of global pandemics: A reality or a mirage the Kenyan perspective? 22 % of Kenyans stay in towns and the inhabitants in these cities continue to grow at the rate of 4.2 % annually. This growth rate has outstripped the supply of housing units built. For instance, Nairobi needs a minimum of 120,000 new houses per annually to satisfy the demand but a paltry 35,000 units are constructed annually. The excess demand is likely to continue pushing the housing prices beyond the reach of many Kenyans. Studies conducted in Kenya on housing prices focused on non-macroeconomic determinants and more importantly none of the studies globally envisaged how global pandemics can influence housing prices. Therefore, the influence of global pandemics like Corona Virus Disease (COVID-19) and macroeconomic factors on housing prices in Kenya remains unknown.
 Study Design: Correlational research design.
 Methodology: The study employed unrestricted Vector Autoregressive analysis involving quarterly time series from quarter 1 of 2014 to quarter 1 of 2020 with a dummy variable measuring the influence of COVID-19.
 Results: Results indicated that the total money supply had a positive influence on inflation that was highly influenced by extended broad money.
 Conclusion: From the results, it was concluded that some macroeconomic factors, time trends and global pandemics like COVID-19 influence housing prices in Kenya. Professional, administrative and support services, time trend, transport and storage, information and communication, real estate and housing prices at lag 1 increased housing prices in Kenya by 0.41%, 0.41%, 0.94%, 0.37% and 0.59% respectively given unrestricted VAR coefficients and t-statistics of 0.41(4.184), 1.27 (9.862), 0.19 (2.740), 0.94 (10.178) and 0.59 (6.055) for the variables. Housing prices at lag 1 and 4, COVID-19, other services and tax on products reduced housing prices in Kenya by 0.26%, 0.99%, 3.29%, 1.01% and 0.05% respectively given unrestricted VAR coefficients and t-statistics of -0.26(-2.366), -0.99 (-8.770), -3.29 (-4.550), -1.01 (-6.568) and -0.05 (-2.807) for the variables respectively. Economic growth, financial and insurance activities and previous housing prices at lag 5 had no influence on housing prices in Kenya.

Cash-flow or return predictability at long horizons? The case of earnings yield

Abstract We examine the predictive ability of the aggregate earnings yield for both market returns and earnings growth by estimating variance decompositions at multiple horizons. Based on weighted long-horizon regressions, we find that most of the variation in the earnings yield is due to return predictability, with earnings growth predictability assuming a minor role. However, by using implied estimates from a first-order restricted VAR, we find an opposite predictability mix. The inconsistency in results stems from a misspecification of the restricted VAR. Using an unrestricted first-order VAR estimated by OLS, or alternatively, estimating the restricted VAR by the Projection Minimum Distance method, produces long-run variance decompositions that are substantially more similar to the decomposition obtained under the direct method. Hence, earnings yield is not fundamentally different from the dividend yield. These results suggest that the practice of analyzing long-run return and cash-flow predictability from a restricted VAR can be quite misleading.

International interdependency of macroeconomic activities: a multivariate empirical analysis

Given the international interdependency of macroeconomic activities, this study attempts to examine three fundamental questions; (i) what is the time pattern of international interdependency?, (ii) what is the magnitude of international interdependency?, and (iii) which are the major channels through which international interdependency occur?. These questions are investigated among major economies viz. the US, Japan, Germany, China, India and Russia. Employing generalized impulse response of unrestricted VAR model, the Diebold and Yilmaz (Int J Forecast 28(1):57–66, 2012) spillover method and orthogonal impulse response of restricted VAR model, the study found that the global economies are highly interdependent. Therefore, we strongly argue that channels through which interdependency between countries occur are based on economic structure of respective economies. From the findings, it explicitly indicates that if there is rise or fall of macroeconomic activity in any specific economy, it may lead to rise or fall in economic activity of other individual economies. It shows a positive correlationship between the movement of economic actvities among the countries. Given the higher degree of international interdependency, it may sometimes lead to larger cost for the individual economies. Therefore, it suggests that the formulation and implementation of rigorous macroeconomic policies are the need of the hour for respective economies. Considering the gaps in the previous literature, this study tries to contribute to the literature by investigating cross-country time profile of interactions, magnitude of interactions and channels through which interactions occur.

Large-dimensional Dynamic Factor Models: Estimation of Impulse–Response Functions with [formula omitted] cointegrated factors

We study a large-dimensional Dynamic Factor Model where: (i) the vector of factors Ft is I(1) and driven by a number of shocks that is smaller than the dimension of Ft; and, (ii) the idiosyncratic components are either I(1) or I(0). Under (i), the factors Ft are cointegrated and can be modeled as a Vector Error Correction Model (VECM). Under (i) and (ii), we provide consistent estimators, as both the cross-sectional size n and the time dimension T go to infinity, for the factors, the loadings, the shocks, the coefficients of the VECM and therefore the Impulse–Response Functions (IRF) of the observed variables to the shocks. Furthermore, possible deterministic linear trends are fully accounted for, and the case of an unrestricted VAR in the levels Ft, instead of a VECM, is also studied. The finite-sample properties the proposed estimators are explored by means of a MonteCarlo exercise. Finally, we revisit two distinct and widely studied empirical applications. By correctly modeling the long-run dynamics of the factors, our results partly overturn those obtained by recent literature. Specifically, we find that: (i) oil price shocks have just a temporary effect on US real activity; and, (ii) in response to a positive news shock, the economy first experiences a significant boom, and then a milder recession.

Financial stress index, growth and price stability in India: some recent evidence

In this paper, an aggregate financial stress index for India is constructed by taking monthly data from different segments of the financial market like money market, bond market, equity market, foreign exchange market, and the banking sector, for the period March 2007 to December 2016. The interrelationship and feedback effect between financial stress, economic growth and price stability are tested by using correlation and an unrestricted VAR model. The impulse response analysis shows that financial stress leads to a decline in growth after a lag period and a higher growth rate for a longer period of time increases stress in the financial system. The variance decomposition result indicates that the contribution of FSI to the variation of other variables are not much high, whereas other variables can affect FSI to some extent. In the Short-run price stability increases financial stress but in the long run, the result is the opposite.

The Effects of COVID-19 Pandemic on Oil Prices, CO2 Emissions and the Stock Market: Evidence from a VAR Model

This paper examines the impact of COVID-19 pandemic on oil prices, CO2 emissions and stock market volatility over the period January 22, 2020 – March 30, 2020 using an unrestricted VAR. We demonstrate that although the increasing number of COVID-19 infections causeda decrease in the price of crude oil, the negative response of the oil market is short-lived.However, the response of economic activities as measured by CO2 emissions to a shock on COVID-19 infections is negative throughout the forecast period. Also, we find that the current situation created by COVID-19 led appears to have had a stronger impact on equity market volatility than on crude oil prices and CO2 emissions. Lastly, we find that the share of forecast error variance in the level of CO2 emissions is stronger than that of the energy and stock markets. Taken together, our findings shed light on the need for economic intervention to speed up recovery and boost investor’ perception of long-term growth.