The impact of geopolitical risks on connectedness among natural resource commodities: A quantile vector autoregressive approach
| dc.contributor.author | Pinar Evrim-Mandaci | |
| dc.contributor.author | Asil Azimli | |
| dc.contributor.author | Nazif Mandaci | |
| dc.contributor.author | Mandaci, Nazif | |
| dc.contributor.author | Mandaci, Pinar Evrim | |
| dc.contributor.author | Azimli, Asil | |
| dc.contributor.author | Evrim Mandaci, Pınar | |
| dc.date.accessioned | 2025-10-06T17:49:24Z | |
| dc.date.issued | 2023 | |
| dc.description.abstract | This study examines the impact of global geopolitical risk on connectedness among major natural resource commodities. We implemented a Quantile Vector Autoregressive connectedness estimation approach from 5 January 2010 to 3 March 2023 including many geopolitical turbulences such as the Russian-Ukrainian war. We found high connectedness under both extraordinarily high and low return conditions. The extreme return shocks in metals tended to spillover to energy commodities. The spillover index peaked during important economic political and financial developments. In addition geopolitical risk drives connectedness among natural resources commodities under average market conditions. Our results may help investors with portfolio optimization and risk management practices and guide policymakers toward attaining financial market stability. © 2023 Elsevier B.V. All rights reserved. | |
| dc.description.sponsorship | Augmented Dickey-Fuller, (49.00, 86.62) | |
| dc.description.sponsorship | Table 1 reports the daily return series' summary statistics and the unit-root test results. Accordingly, palladium offers the highest mean daily return (0.036%), whereas neutral gas (gas) has the highest negative mean daily return (−0.013%). Given the characteristic of daily returns, the standard deviation values are high, ranging between 1.009 and 3.233, with gold having the lowest and gas having the highest fluctuations. Furthermore, daily price changes are higher for coal (−43.245% and 34.057%) and nickel (−52.288% and 52.226%). It is also important to note that most series are left-skewed with negative skewness values. The value for kurtosis is significantly higher than 3, supporting the results of Jacque-Barra (JB) statistics that the returns are not normally distributed, showing leptokurtic and left-skewed patterns. Finally, the Augmented Dickey-Fuller (ADF) test results indicate stationarity.This section examines the commodity spillover connectedness under extreme market conditions, specifically at the lower tail (at the 5th quantile) and upper tail (at the 95th quantile) of the conditional return distribution. Results reported in Table 3 demonstrate significant differences in the connectedness among the commodity future returns across different quantiles. The TCI reported in Panel A at the 5th percent quantile is 87.19. The TCI reported in the last row of Panel B also shows a high level of connectedness (86.62) at the 95th percent quantile. These findings imply that the connectedness under extreme return conditions is higher than the average total spillover index reported at the median (49.00 in Table 2). The commodities are highly interlinked during the extremely negative and positive return episodes. Another exciting feature of spillover results under extreme conditions is that the autocorrelation effects are lower, and cross-correlation effects are higher for all commodities examined. This shows that cross-asset spillovers increase under extreme market conditions compared with connectedness at the median, as reported in Table 2, implying lower benefits to diversification under extreme return episodes. Accordingly, extreme tail behavior in one commodity is more sensitive to the extreme tail behavior in other commodities. These findings support the previous findings showing higher interlinkages among financial assets in stressful times when the returns are highly negative (e.g., Ang and Bekaert, 2002). Further, consistent with Bouri et al. (2021) related to interlinkages among cryptocurrencies at both tails of the conditional return distribution, we find that TCIs among commodities are significantly higher during extremely low and extremely high returns. Results reported in Fig. 1 support the premise that the total spillovers index is equally higher at both tails compared to the median; there is a systematic connectedness structure overall quantiles with higher TCI at both extremes. | |
| dc.identifier.doi | 10.1016/j.resourpol.2023.103957 | |
| dc.identifier.issn | 03014207 | |
| dc.identifier.issn | 0301-4207 | |
| dc.identifier.issn | 1873-7641 | |
| dc.identifier.scopus | 2-s2.0-85165541559 | |
| dc.identifier.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85165541559&doi=10.1016%2Fj.resourpol.2023.103957&partnerID=40&md5=b162e291e32579807725c69d1bcd1bb6 | |
| dc.identifier.uri | https://gcris.yasar.edu.tr/handle/123456789/8425 | |
| dc.identifier.uri | https://doi.org/10.1016/j.resourpol.2023.103957 | |
| dc.language.iso | English | |
| dc.publisher | Elsevier Ltd | |
| dc.relation.ispartof | Resources Policy | |
| dc.rights | info:eu-repo/semantics/closedAccess | |
| dc.source | Resources Policy | |
| dc.subject | Commodities, Connectedness, Geopolitical Risk, Natural Resources, Commerce, Financial Data Processing, Financial Markets, Investments, Risk Management, Auto-regressive, Commodity, Condition, Connectedness, Economic Development, Energy Commodity, Estimation Approaches, Financial Development, Geopolitical Risks, Return Shock, Natural Resources, Commodity Market, Economic Development, Financial Market, Financial System, Geopolitics, Natural Resource, Russian Federation, Ukraine | |
| dc.subject | Commerce, Financial data processing, Financial markets, Investments, Risk management, Auto-regressive, Commodity, Condition, Connectedness, Economic development, Energy commodity, Estimation approaches, Financial development, Geopolitical risks, Return shock, Natural resources, commodity market, economic development, financial market, financial system, geopolitics, natural resource, Russian Federation, Ukraine | |
| dc.subject | Commodities | |
| dc.subject | Connectedness | |
| dc.subject | Natural Resources | |
| dc.subject | Geopolitical Risk | |
| dc.title | The impact of geopolitical risks on connectedness among natural resource commodities: A quantile vector autoregressive approach | |
| dc.type | Article | |
| dspace.entity.type | Publication | |
| gdc.author.id | Mandacı, Nazif/0000-0003-0483-4005 | |
| gdc.author.id | Azimli, Asil/0000-0003-3547-6263 | |
| gdc.author.scopusid | 44861244200 | |
| gdc.author.scopusid | 55253691500 | |
| gdc.author.scopusid | 57209330284 | |
| gdc.author.wosid | EVRIM MANDACI, PINAR/A-3090-2019 | |
| gdc.author.wosid | Azimli, Asil/AAA-3933-2020 | |
| gdc.author.wosid | Mandacı, Nazif/KVX-8842-2024 | |
| gdc.bip.impulseclass | C4 | |
| gdc.bip.influenceclass | C5 | |
| gdc.bip.popularityclass | C4 | |
| gdc.coar.type | text::journal::journal article | |
| gdc.collaboration.industrial | false | |
| gdc.description.department | ||
| gdc.description.departmenttemp | [Mandaci, Pinar Evrim] Dokuz Eylul Univ, Dept Business Adm, Fac Business, Tinaztepe Campus, TR-35390 Buca Izmir, Turkiye; [Azimli, Asil] Cyprus Int Univ, Dept Accounting & Finance, Via Mersin 10, Haspolat, North Cyprus, Turkiye; [Mandaci, Nazif] Yasar Univ, Dept Int Relat, Univ St,N 37-39, Bornova, Izmir, Turkiye | |
| gdc.description.publicationcategory | Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı | |
| gdc.description.startpage | 103957 | |
| gdc.description.volume | 85 | |
| gdc.description.woscitationindex | Social Science Citation Index | |
| gdc.identifier.openalex | W4385093991 | |
| gdc.identifier.wos | WOS:001066867500001 | |
| gdc.index.type | Scopus | |
| gdc.index.type | WoS | |
| gdc.oaire.diamondjournal | false | |
| gdc.oaire.impulse | 15.0 | |
| gdc.oaire.influence | 2.7703289E-9 | |
| gdc.oaire.isgreen | false | |
| gdc.oaire.popularity | 1.2933101E-8 | |
| gdc.oaire.publicfunded | false | |
| gdc.openalex.collaboration | International | |
| gdc.openalex.fwci | 16.1841 | |
| gdc.openalex.normalizedpercentile | 0.99 | |
| gdc.openalex.toppercent | TOP 10% | |
| gdc.opencitations.count | 11 | |
| gdc.plumx.crossrefcites | 11 | |
| gdc.plumx.mendeley | 28 | |
| gdc.plumx.scopuscites | 27 | |
| gdc.scopus.citedcount | 27 | |
| gdc.virtual.author | Mandaci, Pinar Evrim | |
| gdc.virtual.author | Mandaci, Nazif | |
| gdc.wos.citedcount | 22 | |
| person.identifier.scopus-author-id | Evrim-Mandaci- Pinar (44861244200), Azimli- Asil (57209330284), Mandaci- Nazif (55253691500) | |
| project.funder.name | Table 1 reports the daily return series' summary statistics and the unit-root test results. Accordingly palladium offers the highest mean daily return (0.036%) whereas neutral gas (gas) has the highest negative mean daily return (−0.013%). Given the characteristic of daily returns the standard deviation values are high ranging between 1.009 and 3.233 with gold having the lowest and gas having the highest fluctuations. Furthermore daily price changes are higher for coal (−43.245% and 34.057%) and nickel (−52.288% and 52.226%). It is also important to note that most series are left-skewed with negative skewness values. The value for kurtosis is significantly higher than 3 supporting the results of Jacque-Barra (JB) statistics that the returns are not normally distributed showing leptokurtic and left-skewed patterns. Finally the Augmented Dickey-Fuller (ADF) test results indicate stationarity.This section examines the commodity spillover connectedness under extreme market conditions specifically at the lower tail (at the 5th quantile) and upper tail (at the 95th quantile) of the conditional return distribution. Results reported in Table 3 demonstrate significant differences in the connectedness among the commodity future returns across different quantiles. The TCI reported in Panel A at the 5th percent quantile is 87.19. The TCI reported in the last row of Panel B also shows a high level of connectedness (86.62) at the 95th percent quantile. These findings imply that the connectedness under extreme return conditions is higher than the average total spillover index reported at the median (49.00 in Table 2). The commodities are highly interlinked during the extremely negative and positive return episodes. Another exciting feature of spillover results under extreme conditions is that the autocorrelation effects are lower and cross-correlation effects are higher for all commodities examined. This shows that cross-asset spillovers increase under extreme market conditions compared with connectedness at the median as reported in Table 2 implying lower benefits to diversification under extreme return episodes. Accordingly extreme tail behavior in one commodity is more sensitive to the extreme tail behavior in other commodities. These findings support the previous findings showing higher interlinkages among financial assets in stressful times when the returns are highly negative (e.g. Ang and Bekaert 2002). Further consistent with Bouri et al. (2021) related to interlinkages among cryptocurrencies at both tails of the conditional return distribution we find that TCIs among commodities are significantly higher during extremely low and extremely high returns. Results reported in Fig. 1 support the premise that the total spillovers index is equally higher at both tails compared to the median, there is a systematic connectedness structure overall quantiles with higher TCI at both extremes. | |
| publicationvolume.volumeNumber | 85 | |
| relation.isAuthorOfPublication | f4371eab-1073-4761-acbd-e8721f8089c1 | |
| relation.isAuthorOfPublication | 4363c7aa-3999-460d-a92f-e97cc93ae5de | |
| relation.isAuthorOfPublication.latestForDiscovery | f4371eab-1073-4761-acbd-e8721f8089c1 | |
| relation.isOrgUnitOfPublication | ac5ddece-c76d-476d-ab30-e4d3029dee37 | |
| relation.isOrgUnitOfPublication.latestForDiscovery | ac5ddece-c76d-476d-ab30-e4d3029dee37 |