Vector Autoregression Research Articles

On a Partially Non-Stationary Vector AR Model with Vector GARCH Noises: Estimation and Testing

On a Partially Non-Stationary Vector AR Model with Vector GARCH Noises: Estimation and Testing

Evidence on an endogenous growth model with public R&D

ABSTRACT The empirical investigation of properties of an endogenous growth model by Huang, Lai, and Peretto ([2023]. “Public R&D, Private R&D and Growth: A Schumpeterian Approach.” public.econ.duke.edu mimeo) in this paper confirms important assumptions and results of the model for OECD countries. Labour-augmenting technical change is enhanced through private and public R&D stocks in fully modified and dynamic OLS mean-group and pooled mean-group estimations, also when adding the number of enterprises. The CES spillover functions in the growth model’s functions for R&D stock dynamics are supported through nonlinear estimation under the assumptions of identical or different spillover parameters for private and public R&D. We suggest strong public-to-private spillovers and weak private-to-public spillovers as well as high elasticities of substitution for private–public R&D stocks for private R&D processes and low CES for public R&D processes. We confirm the existence of private–public researcher interaction effects in the private R&D knowledge growth function and provide tentative evidence for the linear model relation between public researchers and firm-level R&D and the hump-shaped relation between public and private researchers (both as % labour force). A vector-autoregressive (VAR) panel model in growth rates produces results, which are in accordance with the impact of public R&D cuts on the steady state and the transitional dynamics of the HLP model.

Analysing the Nexus: Stock Indices and Cryptocurrencies During the Conflict Between Russia and Ukraine

Background: The global economy faced increased instability due to the simultaneous occurrence of two major events, the global pandemic in 2020 and the Russian-Ukrainian conflict in 2022, which impacted the financial markets. Purpose: This paper aimed to evaluate the comovements between the USA (S&P 500), Germany (DAX 30), France (CAC40), Japan (Nikkei 225), Canada (TSX), Russia (MOEX) and Ukraine (PFTS) stock markets and the cryptocurrencies Bitcoin (BTC), Ethereum (ETH), Litcoin (LTC) Dash (DASH/USD), Ripple (XRP) DigiByte (DGB) and Nem (XEM), from February 24, 2022, to April 12, 2023 Methods: The approach to our research question will involve using the causality econometric model, Granger SVAR (Vector Autoregressive). Results: The results showed that stock indices and digital currencies show sharp structural breaks, and not all markets influence cryptocurrencies. The MOEX stock market affects the price formation of BTC, ETH, DGB, XEM, and XRP, while the DAX 30 stock index impacts ETH, LTC, DASH, DGB, and XEM. The Ukraine market (PFTS) influences ETH, but the other stock markets do not influence any of the cryptocurrencies analysed. Conclusion: Investors, policymakers, and other participants operating in the digital currency markets can find valuable information in the study's conclusions when seeking to rebalance their portfolios.

Population Aging, Housing Price, and Household Consumption

Population aging and housing price can affect household consumption, and population aging can indirectly affect household consumption through housing price. This study uses panel data from 31 provinces and cities in China between 2001 and 2018 to construct a panel vector autoregression model for analyzing the impact of population aging and housing price on household consumption level. We determine that population aging and housing price positively impact household consumption level, and population aging also has a positive effect on housing price. The mediating effect model using time series data from 2001 to 2018, with housing price as the mediating variable reveals that population aging directly increases household consumption level, and indirectly increases it through housing price. Finally, this study proposes suggestions such as optimizing the population structure, improving the housing system, and advancing the development of the older adult industry.

Linkage between Crude Oil, Gold, INR-USD Reference Rate and Sensex amidst Covid-19 Pandemic: A Cointegration and Causality Analysis

The primary purpose of this paper is to analyse the shortrun and long-run association between crude oil prices, gold prices, USD-INR reference rate, and stock market index across the COVID-19 pandemic. We have employed Johansen's cointegration test to measure the long-run association among the variables. To measure the short-run association of variables, we employed vector autoregression, and then, to find the direction of causality, we used the Granger causality test. The cointegration test results show that variables are not cointegrated in the precrisis as well as during the crisis period; hence, there is no long-run association among the variables. The VAR results show that gold price was influenced by the stock market index in the pre-crisis era; however, during the crisis, no association was observed. Similar results can also be found in crude oil, USD-INR reference rate, gold price, and stock market. The results also show that the relationships among the variables are dynamic and ever-changing in the preCOVID and during the COVID-19 pandemic.

Vector Autoregressive Model Using Non-Normal Copula Errors

This article focuses on a bivariate vector autoregressive model of lag 1 with non-normal errors. We propose a bivariate error distribution using two identical marginals through a copula function. The copula used for the study is the Farlie-Gumbel-Morgenstern copula, which is considered to be one of the efficient class of copula in describing the dependence between two random variables. The model parameters are estimated using the method of inference functions for margins, and the finite sample properties of the model is illustrated through simulation studies. A real life example is considered to illustrate the applications of the proposed model. The adequacy of the copula model is examined using the two sample version of kolmogorov smirnov test statistic.

Geopolitical risk, supply chains, and global inflation

AbstractThis paper investigates the effects of global geopolitical risks and global supply chain pressures on global inflation for the monthly period of 1999M1–2022M12. The investigation is based on a structural vector autoregression model, where the effects of global oil prices and global monetary policy are controlled for. Four alternative measures of inflation are used, including headline, core, food, and energy inflation. The empirical results show that disruptions in global supply chains are the main drivers of global inflation in the long run as the corresponding shocks explain the lion's share of volatilities in headline inflation (by 32%), core inflation (by 30%), and food inflation (by 22%), followed by oil price shocks and policy rate shocks. In comparison, energy inflation is explained the most by oil price shocks (by 55%) followed by supply chain shocks and policy rate shocks. Positive supply chain pressure and oil price shocks have positive and statistically significant effects on headline inflation even after five years, whereas positive policy rate shocks have negative and statistically significant effects on headline inflation in the long run. In contrast, positive shocks to geopolitical risk result in higher headline inflation only up to one year, with insignificant effects in the long run. Several policy implications follow.

The Implementation of Central Bank Policy during the Economic Crisis in Lebanon

During the last four years Lebanon has been facing a severe crisis that have no way out. It has challenged all the potential of the authorities dealing with financial, social and economic sector circumstances. Recently this economic crisis is still going up in higher frequency and more effectively on the Lebanese people. During this crisis central bank of Lebanon implemented its own monetary policy that is represented by expansionary monetary policy. The aim of this paper is to detect the impact of central bank implemented monetary policy during the economic crisis in Lebanon using money supply and open market operations tools in both short and long run on coincident index and consumer price index. In this consideration we perform monthly data covering the period January 2017 to June 2022 by applying VAR model. The empirical results shows that money supply and open market operation have no impact on coincident index and consumer price index in the long run, in addition they do not cause them in the short run. However the impulse response function show some reactions of coincident index and consumer price index to shocks in money supply and open market operations.

Tractable bayesian estimation of smooth transition vector autoregressive models

Abstract We develop a tractable way of estimating the parameters ruling the nonlinearity in the popular Smooth Transition VAR model, and identify structural shocks using external instruments. This jointly offers an alternative to the option of identifying shocks recursively and calibrating key parameters. In an illustration, we show that monetary policy shocks generate larger effects on economic activity during economic expansions compared to economic recessions. We then document that calibrating rather than estimating the parameters ruling the nonlinearity of the model can lead to values for which the key results are lost. This suggests caution in the calibration of these parameters.

CO2 Emissions Mitigation in ASEAN Countries: The Role of Renewable Energy Use and Financial Development

The nations comprising the Association of Southeast Asian Nations (ASEAN) face a pressing need to enhance environmental sustainability, given their substantial reliance on fossil fuels, contributing to 5.2% of global CO2 emissions. Thus, understanding the pivotal role of renewable energy in this context is paramount. This research employs a panel vector autoregression method to analyze the interplay of financial development, renewable energy, and CO2 emissions across ASEAN nations from 1990 to 2020. Findings reveal that while renewable energy consumption shows an insignificant impact on carbon emissions, financial development, and economic growth significantly and positively influence emissions. Conversely, labor exhibits a notable negative correlation with CO2 emissions. Moreover, a bidirectional relationship exists between financial development and economic growth, as well as between labor and GDP. Additionally, unidirectional links are observed from capital formation to CO2 emissions, from renewable energy utilization to fossil fuel dependency, and from renewable energy usage to capital formation. These outcomes underscore the inadequacy of current financial systems in fostering environmental sustainability, highlighting the urgent need for integrating environmental considerations into their operations.

Statistical analysis of gold production in South Africa using ARIMA, VAR and ARNN modelling techniques: Extrapolating future gold production, Resources–Reserves depletion, and Implication on South Africa's gold exploration

South African gold mining is one of the main contributors to the country's economic growth and employment creation. Assessment of future gold production is pivotal for positioning the country's supply security and contribution to the projected global demand. In this study, we modelled and forecasted South African and global gold production trends (1990–2022) using the Autoregressive Integrated Moving Average (ARIMA), Vector Autoregressive (VAR), and Autoregressive Neural Network (ARNN) techniques. The study also evaluated the impact of annual productions on the collated reserves (2008–2021) and the contribution of resource–reserve conversion in sustaining gold reserves within the South African goldfields. The ARIMA (1,1,0) model, based on root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), outperformed both ARNN (1,1) and VAR (2) methods. The forecast of the ARIMA model showed an increase in South Africa's gold production, with a compounded annual growth rate (CAGR) of ∼0.6%. The model also predicted an increase in global gold production, with CAGR of ∼2.2%. The analyses of reserves–production trends showed that most South African mining companies have been depleting their gold reserves due to production and were minimally converting resources to reserves, averaging at ∼2%/year from 2017 till present. The minimal conversion may also reflect the challenging mining conditions (such as mining depths and underground flooding) and the mining industry's socio-political dynamics. We demonstrated that, if the resource–reserve conversion factor is maintained (at ∼2%/year and above), the forecasted production may sustain the current reserves (at > 2000 t Au excluding tailings) but deplete ∼27 % of the resources to reserves by 2040. However, if the conversion factor continues to decline (to <1 %), the forecasted gold production will reduce the reserves to < 500 t Au with ∼3% of the resources converted to reserves.

Determinant of Market Liquidity in Indonesia

This research aims to explore the Market volatility, Market return, exchange rate and foreign stock buying, on market liquidity in the Indonesia Stock Exchange. This research use VAR (Vector Autoregression) to estimate coefficient model. Monthly data is used for period of January 2010 to December 2023. This research found that Market Liquidity lag – 1 and Lag-2 positively significant impact on Market Liquidity. Market Return negatively significant influence on Market Liquidity. Exchange Rate negatively significant impact on market liquidity. Pandemic era or Covid-19 positively significant affected on market liquidity. IDX issued a rule to reduce 1 lot trading to become 100 stocks which is positively significant influence on Market Liquidity. Stock Volatility and Foreign Buying did not affect on market liquidity.

The effect of structural oil shocks on stock returns of Indian renewable energy companies across market conditions

Purpose This study aims to evaluate the structural oil shocks effect on stock returns of Indian renewable energy companies across market conditions. Design/methodology/approach This study applies the structural vector autoregression model to estimate sources of oil shocks such as oil supply shock, aggregate demand shock and oil price-specific demand shock. In the next step, the panel quantile regression model estimates the effect of these oil shocks on stock return across market conditions. Monthly data are collected from January 2009 to December 2019. All renewable energy companies listed on the National Stock Exchange of India are considered for the analysis. Findings In the whole sample analysis, this study finds that oil shocks negatively affect stock returns in most of the market conditions except oil price-specific demand shock. In sub-groups, oil shocks driven by supply and aggregate demand also negatively affect stock return in most market conditions. This study finds the positive interaction of oil price-specific demand shock. A majority of these positive interactions happen in bearish market conditions. In the whole sample, the asymmetric effects of shocks driven from oil supply and oil price-specific demand are seen in most quantiles or market conditions. At the same time, aggregate demand shock does not affect asymmetrically. In the sub-group analysis, standalone renewable energy companies stock returns are least asymmetrically affected by these oil shocks. The asymmetries of oil supply-driven shock on stock returns of the renewable energy sub-group companies are found in most quantiles. Originality/value First, this is a company-level study of the stock returns response to the structural oil shocks in the renewable energy sector. Second, to the best of the authors’ knowledge, this type of study is the first in the Indian context. Third using panel quantile regression model along with capital asset pricing model framework, the authors investigate these effects across market conditions.

Exploring the nexus of decent work, financial inclusion, and economic growth: A study aligned with SDG 8

We investigate the potential influence of fair employment practices and financial inclusion on economic growth, focusing on aligning our analysis with the United Nations Sustainable Development Goal 8 (SDG 8). Our study encompasses various countries, including 26 low-income, 20 middle-income, and 15 high-income nations. Utilizing a dynamic fixed-effect panel data approach and employing the Generalized Method of Moments from 2010 to 2021, our findings reveal that conducive environments for decent work and enhanced access to digital financial services positively impact economic growth. This synergistic effect of SDG 8 components is particularly pronounced in high-income countries, suggesting the potential for accelerated economic growth rates. Conversely, our analysis finds no discernible impact on the economic growth of low-income countries. For middle-income countries, nuanced coefficient outcomes provide a more intricate understanding of their economic dynamics. Through extensive impulse-response analysis using a Panel Vector Autoregression model, we illustrate the distinct effects of shocks related to decent work and financial inclusion across various income categories. Our results are robust to multiple sensitivity tests, underscoring their reliability and signaling significant policy implications for acknowledging the critical role of decent work and financial inclusion in fostering economic growth.

The role of environmental tax on the environmental quality in EU counties: evidence from panel vector autoregression approach.

This study intends to analyze the influence of environmental taxes on pollution in EU-27 nations. Furthermore, energy from renewable sources consumption and urbanization are employed to clarify CO2 emissions in this study that tests the EKC hypothesis. According to the findings, an increase in environmental taxes reduces CO2 emissions by 0.14%. Also, the data supported the validity of the EKC concept. The findings of the causality test demonstrated that there is a bidirectional causal link between CO2 emissions and environmental taxes. These results reflect that environmental tax revenues contribute to sustainability as an effective policy tool in EU countries. Policies regarding environmental tax enforcement come to the fore in terms of both keeping the balance in economic activities and serving sustainability.

Systematic evaluation of integration between China's digital economy and sports industry: Two-stage grey relational analysis and vector autoregressive model.

The development of the digital economy constitutes a key component of China's endeavors to advance towards "Digital China." The sports industry functions as a new catalyst for high-quality economic growth. This study systematically evaluated the integration between these two sectors. First, we conducted two levels of grey relational analysis to assess their integration between 2016 and 2021. Second, we conducted a VAR analysis to determine whether their integration between 2009 and 2021 represents a causal relationship. At the macro level, the grey relational analysis reveals that the sports industry (grade = 0.770) ranked second among China's eight key economic sectors in terms of digital economy integration. At the meso level, a wide variation (ranging from 0.606 to 0.789) existed in the grade of integration between the digital economy and the sub-sectors of the sports industry. According to the VAR model, the digital economy does not Granger cause (p = 0.344) the growth of the sports industry. This study yielded two added values to the existing literature: First, there exists a sectoral imbalance in the digitization process; second, the explosive growth of the sports industry was not primarily caused by the digital economy. Accordingly, the "sports + digital" complex is still in the first wave of technological integration. We propose three policy recommendations, namely, sectoral synergistic development, overtaking via esports IP, and new economy and new regulation. Collectively, these findings provide updated insights for the digital transformation towards "building a leading sports nation" and "Digital China."

Semantic Segmentation for Change Detection in Satellite Imaging

Change detection is a common and actual problem in the field of remote sensing. The classical approaches using raw pixel information are very sensitive to noise. In this study we propose the usage of additional semantic information for change detection. We use the semantic segmentation methods like geospatial Segment Anything Model and encoder based U-Net to evaluate the predictions and tracing the semantic information as well as raw information in change detection. Later the multidimensional time series data is used via the Vector Autoregression model to predict the future changes in the landscape. The observations which fall out of the prediction interval are considered as the changes in the landscape. The proposed method is evaluated on the dataset of the random locations across the Baltic region. The research is accompanied by the data and reproducible code at Github repository.

Revisiting the role of climate change on crop production: evidence from Mediterranean countries

AbstractThe Mediterranean region is an important agricultural center and is expected to be most affected by climate change due to its location. In this study, the role of climate change on agriculture is examined for eight South European countries on the Mediterranean coast for the period of 1996–2016. Carbon dioxide emissions, temperature and precipitation values are used as the indicators of climate change while cereal production is taken as a proxy for agricultural production. Results of the panel vector autoregression approach reveal that an increase in the carbon dioxide emissions and temperature have a negative impact on the cereal production. However, this effect is statistically significant only in the initial periods. On the other hand, an increase in rainfall has a statistically significant positive effect on crop production only in one period. However, eventually this effect turns to negative as expected, since excessive rainfall has a negative effect on agriculture as much as drought. In conclusion, the importance of bringing the policies and strategies to the forefront in ensuring the sustainability of agriculture and minimizing the negative effects of climate change in the region are discussed.

Exploring inflation dynamics in Canada: A threshold vector autoregressive approach

Given the pivotal role of inflation expectations in contemporary monetary policy, we posit that if monetary policy has effectively influenced inflation expectations, thereby altering the trajectory of total inflation, a structural break in the path of total inflation should be observable. Conversely, if inflation expectations have remained stable and monetary policy has had limited impact, a stable vector autoregressive (VAR) model should adequately describe the path of total inflation. To address these hypotheses, a non-linear specification of a threshold vector autoregressive (TVAR) model is employed, offering a comprehensive analytical framework for the examination of these dynamics.

Machine learning-enabled weather forecasting for real-time radioactive transport and contamination prediction

Effective responses to nuclear accidents require accurate and fast predictions of radionuclide transport in the atmosphere. However, real-time forecasting has been challenging, since predicting the transport requires meteorological forecasting, which relies on computationally intensive numerical weather simulations. To address this challenge, we have developed statistical emulators for the Weather Research and Forecasting model (WRF) to forecast wind and temperature fields around the source location. The emulator is constructed based on pre-computed WRF simulation results and their boundary conditions from global circulation models (GCM), enabling the rapid prediction of local-scale wind/temperature fields by ingesting the GCM-predicted boundary conditions available online for the next several days. We have explored two methods: (1) deep neural network and (2) a vector autoregressive model with exogenous variables (VARX). A Gaussian plume model is then used to take these field as input and to predict the distribution of radioactive material/dose in the environment. The results show that both emulators are capable of forecasting the wind field for several days in the future in real-time, while not compromising accuracy. The autoregressive model has shown a strong performance of predicting the temperature field, owing to the strong temporal autocorrelations in the temperature data. This framework offers a powerful tool for informed decision-making during emergencies such as enabling meteorology-informed timing of containment venting, and real-time guidance of evacuation strategies.