Article in International Journal of Entrepreneurship · December 2020 doi: 10. 4018/ijegcc. 2020070105 citations 11 reads 286 João Manuel Pereira Polytechnic Institute of Setúbal Instituto Politécnico de Lisboa 95

Figure 2. Evolution in % of the differences, of the eight financial markets, in the period from 31/12/2019 to 23/07/2020

Table 2. Descriptive statistics, on the returns, of the eight European financial markets from 31/12/2019 to 23/07/2020

	ATG	CAC 40	DAX 30	FTSE 100	FTSE MID	IBEX 35	ISEQ	PSI 20
Mean	-0.001538	-0.000477	0.000467	-0.000692	-0.000197	-0.001036	4.30E-05	-0.000452
Std.Dev.	0.027084	0.021336	0.020510	0.019336	0.023779	0.022028	0.021069	0.017225
Skewness	-1.521235	-1.456190	-1.299364	-1.177540	-2.900081	-1.828632	-1.076405	-1.521948
Kurtosis	10.92967	11.74239	14.13378	11.50055	24.02861	15.09658	8.213614	13.94787
Jarque-Bera	598.1297***	704.0564***	1083.840***	645.1387***	3945.543***	1324.203	*** 263.8107***	1070.629***
Observations	199	199	199	199	199	199	199	199

Source: Own elaboration

Table 3. Stationary test by Levin, Lin, and Chu (2002), applied to the eight European financial from 12/31/2019 to 07/23/2020

Method	Statistic	Prob.**		Cross-Sections	Obs
Null:Unitroot(assumescommonunitrootprocess)
Levin,Lin&Chut*	-42.3461	0.0000	8		1582
Null:Unitroot(assumesindividualunitrootprocess)
Im,PesaranandShinW-stat	-36.1778	0.0000	8		1582
ADF-FisherChi-square	726.350	0.0000	8		1582
PP-FisherChi-square	852.623	0.0000	8		1584

Source: Own elaboration

Note: ** Probabilities for Fisher tests are computed using an asymptotic Chi-square distribution. All other tests assume asymptotic normality.

thenumberoflagstoincludeinthecausalitytests,theauthorsusedtheHQ(Hannan-Quinninformation criterion)criterionthatsuggeststwolags(seetable7).

Figure4showstheautocorrelationintheresidueswithtwolagsinthefinancialmarkets,which isnotobservedinthecaseoftheCOVID-19dataseries(confirmedcasesanddeaths).Asmaller numberoflagsincreasesthedegreesoffreedom.Amoresignificantnumberoflagsdecreasesthe problemsofautocorrelation.Beingthisthecase,theauthorsestimatedaVARwithfourlagsforthe dataseriesoftheEuropeanfinancialmarkets.

InTable8,onecanobservetheresultsoftheVARResidualSerialCorrelationLMTests.The authorsperformedaVARestimationwithfourlags,followedbytheautocorrelationtestatfivelags. Thenullhypothesiswasnotrejected,whichprovesthatthemodeldoesnotpresentautocorrelation inresidues,indicatingthatitpresentsarobustestimation.

TheresultsofGranger’scausalitytestsareshownintable9,bothreferringtoCOVID-19data (confirmedcasesanddeaths)andthefinancialmarketsunderanalysis.Thecausalitytestsshowtwo bi-directionalcausalrelationshipsbetweenconfirmedcasesanddeathsfromtheCOVID-19virus.

However,therewerenoshocksbetweenCOVID-19data(casesanddeaths)andfinancialmarkets. Thefactthattheevolutionoftheglobalpandemic(COVID-19)didnotcauseshocksinEurope’s financialmarketsduringthisperiodofanalysisdoesnotvalidatethefirstresearchquestioninour studyand,therefore,contradictstheresultsoftheauthorAshraf(2020),whoobservedthatthestock marketsreactednegativelytothegrowthofconfirmedcasesofCOVID-19.Inotherwords,thereturns onthestockmarketsdecreasedasthenumberofconfirmedcasesincreased.

Table10depictstheresultsoftheexponentsDetrendedFluctuationAnalysis(DFA),andone canobservethatEurope’sfinancialmarketsshowsignificantlong-termmemories,rangingfrom 0.61-0.73.Thesefindingsshowthatpricesdonotfullyreflecttheinformationavailableandthat changesinpricesarenoti.i.d.Thissituationhasimplicationsforinvestors,sincesomereturnscan

Table 4. ADF Stationarity test - Fisher Chi-square, applied to eight European Markets in the period from 31/12/2019 to 23/07/2020

	Method		Statistic	Prob.**
ADF-F	isherChi-square		726.350	0.0000
ADF	-ChoiZ-stat		-25.7229	0.0000
Inte	rmediateADFtestresultsUNTITLED
Series	Prob.	Lag	Max Lag	Obs
ATG	0.0000	0	14	198
CAC40	0.0000	0	14	198
DAX30	0.0000	0	14	198
FTSE100	0.0000	0	14	198
FTSEMID	0.0000	1	14	197
IBEX35	0.0000	1	14	197
ISEQ	0.0000	0	14	198
PSI20	0.0000	0	14	198

Source: Own elaboration

Note: ** Probabilities for Fisher tests are computed using an asymptotic Chi-square distribution. All other tests assume asymptotic normality.

beexpected,creatingopportunitiesforarbitrageandabnormalearnings,contrarytoassumedby thehypothesesofrandomwalkandinformationalefficiency,thusvalidatingoursecondresearch question.TheseresultsarecorroboratedbytheauthorsAggarwal(2018),Rehman,Chhapra,Kashif, andRehan(2018),Pernagallo,andTorrisi(2019),whohighlightlong-termmemoriesofthereturns onthestockmarkets.

Table 5. PP Stationarity test - Fisher Chi-square, applied to the eight European financial markets in the period from 31/12/2019 to 23/07/2020

Method			Statistic		Prob.**
PP-FisherChi-square			852.623		0.0000
PP-ChoiZ-stat			-28.2955		0.0000
IntermediatePhillips-PerrontestresultsUNTITLED
Series	Prob.	Bandwidth		Obs
ATG	0.0000	4.0		198
CAC40	0.0000	6.0		198
DAX30	0.0000	6.0		198
FTSE100	0.0000	5.0		198
FTSEMID	0.0000	7.0		198
IBEX35	0.0000	6.0		198
ISEQ	0.0000	6.0		198
PSI20	0.0000	6.0		198

Source: Own elaboration

Note: ** Probabilities for Fisher tests are computed using an asymptotic Chi-squar distribution. All other tests assume asymptotic normality.

Figure 3. Stability tests of the residues of the eight European financial markets, in the period from 31/12/2019 to 23/07/2020

Table 6. Unit root tests with structural breaks by Clemente et al. (1998), in returns, referring to the eight financial markets in Europe, in the period from 12/31/2019 to 07/23/2020

Index	t-Stat	Break Date
ATG	-12.57(0)***	05/02/2020
CAC40	-13.53(0)***	31/03/2020
DAX30	-12.82(0)***	10/03/2020
FTSE100	-13.91(0)***	12/03/2020
FTSEMID	-14.61(0)***	17/03/2020
IBEX35	-13.12(0)***	02/04/2020
ISEQ	-12.46(0)***	19/03/2020
PSI20	-13.19(0)***	26/03/2020

Source: Own elaboration

Note: Lag Lenght (Automatic Length based on SIC). Break Selection: Minimize Dickey-Fuller t-statistic. The lateral values in parentheses refer to lags. ***. **. *. represent significance at 1%. 5% and 10%. respectively.

Table 7. VAR Lag Order Selection Criteria

Lag	LogL	LR	FPE	AIC	SC	HQ
2	5552.948	423.7915	2.41e-37	-55.94710	-52.37130	-54.49874*

Source: Own elaboration

Figure 4. VAR Residuals Test

Table 8. VAR Residual Serial Correlation LM Tests

Lag	LRE* stat	df	Prob.	Rao F-stat	df	Prob.
1	205.3774	100	0.0000	2.165009	(100,978.4)	0.0000
2	132.8881	100	0.0155	1.350990	(100,978.4)	0.0158
3	165.5520	100	0.0000	1.710690	(100,978.4)	0.0000
4	134.4604	100	0.0123	1.368044	(100,978.4)	0.0125
5	103.2714	100	0.3913	1.034586	(100, 978.4)	0.3935

Source: Own elaboration

Table 9. Granger Casuality Tests / Block Exogeneity Wald Tests, in the Full Period

	COVID-19 (casos)	COVID-19 (mortes)
COVID-19(casos)	********	3.84(4)***
COVID-19(mortes)	16.02(4)***	********
ATG	0.02(4)	0.06(4)
CAC40	0.11(4)	0.10(4)
DAX30	0.08(4)	0.04(4)
FTSE100	0.13(4)	0.09(4)
FTSEMID	0.10(4)	0.02(4)
IBEX35	0.12(4)	0.04(4)
ISEQ	0.28(4)	0.17(4)
PSI20	0.34(4)	0.18(4)

Source: Own elaboration

Note: The markets in columns “cause” the markets in line. The lateral values in parentheses refer to lags. ***. **. *. represent significance at 1%. 5% and 10%. respectively.

Table 10. DFA exponent for return. The values of the linear adjustments for αDFA always had R² > 0.99

Index		DFA Exponent (COVID-19)
ATG	0.68≌0.0125
CAC40	0.68≌0.0005
DAX30	0.72≌0.0017
FTSE100	0.61≌0.0035
FTSEMID	0.70≌0.0032
IBEX35	0.69≌0.0030
ISEQ	0.69≌0.0011
PSI20	0.73≌0.0025

Source: Own elaboration

Note: The hypotheses are H₀ α = 0.5 and H₁ α ≠ 0.5

CONCLUSION

Knowinghowsensitivityandvulnerablefinancialmarketsaretounexpectedstructuralandcontextual changesisarelevantmatterofstudy.Byunderstandingthecausalrelationshipbetweenthesechanges andthebehaviorofthesemarkets,itispossibletocreate,inadvance,mitigationmechanismsthat cancontributeinsomewaytoreducethepossibledevastatingeffectsontheinterestsofthevarious stakeholders.DuetothedisruptiveandsevereeffectofCOVID-19ontheeconomy,itisvitaltostudy thisphenomenonwiththesamepurposeinmind.

Therefore,theaimofthisstudywastotestwhethertheevolutionofCOVID-19(confirmed casesanddeaths)causedshocksinthestockmarketsofGreece(ATG),France(CAC40),Germany

(DAX30),UnitedKingdom(FTSE100),Italy(FTSEMID),Spain(IBEX35),Ireland(ISEQ),and Portugal(PSI20),intheperiodfromDecember31,2019,toJuly23,2020.Theunderlyinganalysis hasbeenconductedbasedontworesearchquestionstoconfirmthispossibility:Doestheincrease inthenumberofCOVID-19casesanddeathsshockEurope’sfinancialmarkets?Ifso,couldthe presenceoflongmemoriescauseincreasedlevelsofarbitration?

Sincetheunderlyingresearchproblempertainstoarecentphenomenon,theCOVID-19outbreak, theliteraturereviewisbasedonrecentarticlespublishedin2020regardingtheglobalpandemic, withaconsiderableABSRanking.Basedonthisfact,theauthorsunderstandthatitisnotpossible toprovidealiteraturereviewthatgoesbeyondthepresentedtimespanoftheunderlyingresearch topic,astherearenone.NohealthissueliketheCOVID-19pandemichashadintherecentpastsuch adisastrousimpactontheglobaleconomies.

Thefirsttestresultsshowtwobi-directionalcausalrelationshipsbetweenconfirmedcasesand deathsfromtheCOVID-19virus.However,noshocksbetweenCOVID-19data(casesanddeaths) andfinancialmarketswereevidenced.Theseresultssuggestthattheevolutionoftheglobalpandemic (COVID-19)didnotcauseshocksinEurope’sfinancialmarketsduringtheperiodinwhichthe analysiswasperformed.Therefore,basedontheseresults,thefirstresearchquestionwasrejected.

TheresultsoftheobtainedexponentsDetrendedFluctuationAnalysis(DFA)showsignificant longmemoriesintheEuropeanfinancialmarkets,rangingbetween0.61-0.73.Thesefindingsshow thatpricesdonotfullyreflecttheinformationavailableandthatchangesinpricesarenoti.i.d.This situationhasimplicationsforinvestors,sincesomereturnscanbeexpected,creatingopportunities forarbitrageandabnormalearningsthatcontradicttheassumedbythehypothesesofrandomwalk andinformationalefficiency,thusvalidatingthesecondresearchquestion.

Thegeneralconclusiondrawnandsustainedbytheresultsobtainedthroughthetestscarriedout witheconometricandmathematicalmodelsdemonstratesthattheglobalpandemiccreatedturbulence andpessimisminthefinancialmarketsduetouncertaintyandlackofconfidence.Inreturn,these effectsledtosignificantstructuralbreaks.However,whentheauthorstestedtheCOVID-19data series(confirmedcasesanddeaths),thecausalitytestsshowedtwobidirectionalcausalrelationships betweenconfirmedcasesanddeathsresultingfromtheCOVID-19virus.However,therewereno shocksbetweenCOVID-19data(casesanddeaths)andfinancialmarkets,orrathernodirectcausalities betweentheevolutionofCOVID-19andEurope’sfinancialmarkets’profitabilitywasobserved.The memorypropertiesofthesestockmarketsshowpersistenceinreturns,whichrejectsthehypothesis ofmarketefficiencyinitsweakform.Ourstudy’sconclusionsshedlightonmarketregulatorsto ensurebetterinformationamonginternationalfinancialmarkets.Inconclusion,theauthorsbelieve thatinvestorsshoulddiversifytheirportfoliosandinvestinlessriskymarketstomitigateriskand improvetheirportfolios’efficiency.

Finally,thisresearchhasusedgeneralindicesofadailyfrequencytoanalyzetheglobalpandemic’s impactontheanalyzedfinancialmarkets(firstwave).Inthefutureitwouldalsobeinterestingto analyzetheimpactofthesecondwaveofCovid-19ontheanalyzedmarketsusingintradaydata, quotesperminute,aswellasaddingvariablessuchasunemployment,GDP,tocheckwhethermarket fluctuationsaresynchronizedwithmacroeconomicvariables.

REFERENCES

Aggarwal,D.(2018).RandomwalkmodelandasymmetriceffectinKoreancompositestockpriceindex.AfroAsianJ.ofFinanceandAccounting.doi:10.1504/AAJFA.2018.10009906

Akhtaruzzaman,M.,Boubaker,S.,Lucey,B.M.,&Sensoy,A.(2020).IsGoldaHedgeorSafeHavenAsset duringCOVID–19Crisis?SSRNElectronic Journal.10.2139/ssrn.3621358

Ali,M.,Alam,N.,&Rizvi,S.A.R.(2020).Coronavirus(COVID-19)—Anepidemicorpandemicforfinancial markets.Journal of Behavioral and Experimental Finance,27,100341.Advanceonlinepublication.doi:10.1016/j. jbef.2020.100341PMID:32427215

Ashraf,B.N.(2020).Stockmarkets’reactiontoCOVID-19:Casesorfatalities?Research in International Business and Finance,54,101249.Advanceonlinepublication.doi:10.1016/j.ribaf.2020.101249

Clemente,J.,Montañés,A.,&Reyes,M.(1998).Testingforaunitrootinvariableswithadoublechangeinthe

mean.Economics Letters,59(2),175–182.doi:10.1016/S0165-1765(98)00052-4

Dias,R.,daSilva,J.V.,&Dionísio,A.(2019).FinancialmarketsoftheLACregion:Doesthecrisisinfluence thefinancialintegration?International Review of Financial Analysis,63(January),160–173.doi:10.1016/j. irfa.2019.02.008

Dias,R.,Heliodoro,P.,Teixeira,N.,&Godinho,T.(2020).Testing the Weak Form of Efficient Market Hypothesis : Empirical Evidence from Equity Markets.10.11648/j.ijafrm.20200501.14

Dickey,D.,&Fuller,W.(1981).Likelihoodratiostatisticsforautoregressivetimeserieswithaunitroot. Econometrica,49(4),1057–1072.doi:10.2307/1912517

Ferreira,P.,Dionísio,A.,Guedes,E.F.,&Zebende,G.F.(2018).Aslidingwindowsapproachtoanalysethe evolutionofbanksharesintheEuropeanUnion.Physica A,490,1355–1367.doi:10.1016/j.physa.2017.08.095

Gujarati,D.N.(2004).Basic Econometrics.doi:10.1126/science.1186874

Gunay,S.(2020).ANewFormofFinancialContagion:COVID-19andStockMarketResponses.SSRN

Electronic Journal.10.2139/ssrn.3584243

He,Q.,Liu,J.,Wang,S.,&Yu,J.(2020).TheimpactofCOVID-19onstockmarkets.Economic and Political Studies,8(3),275–288.Advanceonlinepublication.doi:10.1080/20954816.2020.1757570

Jarque,C.M.,&Bera,A.K.(1980).Efficienttestsfornormality,homoscedasticityandserialindependenceof regressionresiduals.Economics Letters,6(3),255–259.doi:10.1016/0165-1765(80)90024-5

Kantelhardt,J.W.,Koscielny-Bunde,E.,Rego,H.H.,Havlin,S.,&Bunde,A.(2001).Detectinglongrangecorrelationswithdetrendedfluctuationanalysis.Physica A,295(3–4),441–454.doi:10.1016/S03784371(01)00144-3

Levin,A.,Lin,C.F.,&Chu,C.S.J.(2002).Unitroottestsinpaneldata:Asymptoticandfinite-sampleproperties. Journal of Econometrics,108(1),1–24.Advanceonlinepublication.doi:10.1016/S0304-4076(01)00098-7 Liu,H.,Manzoor,A.,Wang,C.,Zhang,L.,&Manzoor,Z.(2020).TheCOVID-19outbreakandaffected countriesstockmarketsresponse.International Journal of Environmental Research and Public Health,17(8), 2800.Advanceonlinepublication.doi:10.3390/ijerph17082800PMID:32325710

Okorie,D.I.,&Lin,B.(2020).StockmarketsandtheCOVID-19fractalcontagioneffects.Finance Research Letters,101640.Advanceonlinepublication.doi:10.1016/j.frl.2020.101640PMID:32837366

Peng,C.K.,Buldyrev,S.V.,Havlin,S.,Simons,M.,Stanley,H.E.,&Goldberger,A.L.(1994).Mosaic organizationofDNAnucleotides.Physical Review. E,49(2),1685–1689.doi:10.1103/PhysRevE.49.1685 PMID:9961383

Pernagallo,G.,&Torrisi,B.(2019).AnempiricalanalysisonthedegreeofGaussianityandlongmemoryof financialreturnsinemergingeconomies.Physica A,527,121296.Advanceonlinepublication.doi:10.1016/j.

physa.2019.121296 Perron,P.,&Phillips,P.C.B.(1988).TestingforaUnitRootinaTimeSeriesRegression.Biometrika,2(75), 335–346.doi:10.1080/07350015.1992.10509923

Rehman,S.,Chhapra,I.U.,Kashif,M.,&Rehan,R.(2018).Are Stock Prices a Random Walk? An Empirical Evidence of Asian Stock Markets.ETIKONOMI.doi:10.15408/etk.v17i2.7102

RuizEstrada,M.A.,Park,D.,Koutronas,E.,Khan,A.,&Tahir,M.(2020).TheImpactofInfectiousand ContagiousDiseasesandItsImpactontheEconomicPerformance:TheCaseofWuhan,China.SSRNElectronic

Journal.10.2139/ssrn.3527330

Sansa,N.A.(2020).TheImpactoftheCOVID-19ontheFinancialMarkets:EvidencefromChinaandUSA. SSRNElectronic Journal.10.2139/ssrn.3567901

Yan,B.,Logan,S.,Tu,A.,&Zhang,T.(2020).AnalysisoftheEffectofCOVID-19ontheStockMarketand PotentialInvestingStrategies.SSRNElectronic Journal.

Zeren,F.,&Hizarci,A.(2020).The impact of COVID-19 coronavirus on stock markets: Evidence from selected countries.MuhasebeveFinansİncelemeleriDergisi.,doi:10.32951/mufider.706159

Zhang,D.,Hu,M.,&Ji,Q.(2020).FinancialmarketsundertheglobalpandemicofCOVID-19.Finance Research Letters,36,101528.Advanceonlinepublication.doi:10.1016/j.frl.2020.101528PMID:32837360

Rui Manuel Teixeira Santos Dias holds a Ph.D. in Finance from the University of Évora - Institute for Advanced Studies and Research. He is an Adjunct Professor at the Polytechnic Institute of Setúbal, School of Business Administration and Researcher at the University of Évora (CEFAGE).

João Pereira holds a Ph. D. in Management with a specialization in Marketing from the Universidade Aberta and is a visiting professor in his area of specialization at the Higher Institute of Accounting and Administration of the Lisbon Polytechnic Institute, and at the Universidade Aberta, Portugal.

View publication stats