Co esponding au ho : Menaama Amoawah Nk umah
Copy igh © 2025 Au ho (s) e ain he copy igh o his a icle. This a icle is published unde he e ms o he C ea i e Commons A ibu ion License 4.0.
Modeling he impac o da a b eaches on s ock ola ili y using inancial ime se ies
and e en -based isk models
Menaama Amoawah Nk umah *
Depa men o Ma hema ics, Illinois S a e Uni e si y, USA.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
Publica ion his o y: Recei ed on 02 Ap il 2025; e ised on 11 May 2025; accep ed on 13 May 2025
A icle DOI: h ps://doi.o g/10.30574/wja .2025.26.2.1901
Abs ac
Da a b eaches ha e eme ged as c i ical inancial e en s wi h he po en ial o signi ican ly impac in es o con idence,
ma ke s abili y, and s ock p ice ola ili y. As cybe a acks become mo e equen and damaging, he e is a g owing
demand o obus analy ical amewo ks o quan i y hei inancial implica ions. This s udy p esen s a comp ehensi e
app oach o modeling he impac o publicly disclosed da a b eaches on s ock ola ili y using inancial ime se ies
analysis and e en -based isk modeling. The esea ch applies Gene alized Au o eg essi e Condi ional
He e oskedas ici y (GARCH), Exponen ial GARCH (EGARCH), and Vec o Au o eg ession (VAR) models o assess pos -
b each ola ili y pa e ns, spillo e e ec s, and e en lags ac oss di e en indus ies, including echnology, inance, and
e ail. The analysis begins wi h an explo a ion o his o ical s ock pe o mance a ound b each disclosu e windows,
iden i ying ola ili y clus e ing and asymme ic e ec s consis en wi h in es o panic and unce ain y. Using e en
s udy me hodology, abno mal e u ns and ola ili y shocks a e cap u ed and measu ed o e alua e bo h sho - e m and
pe sis en impac s. GARCH and EGARCH models a e used o quan i y ola ili y pe sis ence and asymme ic esponses
o nega i e news, while VAR models assess he spillo e o b each- ela ed shocks ac oss co ela ed secu i ies and
sec o s. Findings e eal ha b each disclosu es ypically esul in sho - e m spikes in ola ili y and nega i e abno mal
e u ns, wi h mo e se e e impac s obse ed in sec o s ha handle sensi i e cus ome da a. Fu he mo e, he ma ke
esponse exhibi s lag e ec s, sugges ing delayed p ice adjus men s as new in o ma ion un olds pos -b each. This s udy
p o ides ac ionable insigh s o ins i u ional in es o s, inancial isk manage s, and egula o s seeking o be e
unde s and and mi iga e cybe secu i y-induced ma ke isk.
Keywo ds: Da a B each; S ock Vola ili y; GARCH; In es o Con idence; E en S udy; Ma ke Risk
1. In oduc ion
1.1. Backg ound: Cybe secu i y and Financial Ma ke In e dependence
In oday's globally in eg a ed economy, inancial ma ke s depend hea ily on digi al in as uc u e o acili a e eal- ime
ansac ions, da a analysis, and egula o y epo ing. This eliance has ele a ed cybe secu i y om a echnical issue o
a cen al pilla o inancial ma ke s abili y [1]. Wi h he p oli e a ion o algo i hmic ading, blockchain applica ions,
and in e connec ed banking sys ems, e en a localized cybe a ack can igge sys emic consequences ha anscend
na ional bo de s [2]. As inancial ins i u ions mig a e hei ope a ions o cloud en i onmen s and le e age open banking
pla o ms, hei a ack su aces con inue o expand, making hem ulne able o inc easingly sophis ica ed h ea s [3].
Cybe inciden s such as he 2017 Equi ax b each and he 2020 Sola Winds a ack ha e demons a ed how
ulne abili ies in digi al ecosys ems can lead o inancial losses, epu a ional damage, and e osion o in es o con idence
[4]. The apid digi iza ion o inancial se ices—accele a ed u he by he COVID-19 pandemic—has deepened he
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2460
in e dependence be ween cybe secu i y esilience and ma ke unc ionali y. Cen al banks and egula o y bodies ha e
ecognized his con e gence, p omp ing a shi in o e sigh amewo ks ha now inco po a e cybe isk as a dimension
o mac op uden ial s abili y [5].
Mo eo e , he ise o decen alized inance (DeFi), high- equency ading, and digi al asse s has in oduced new laye s
o complexi y and ulne abili y. A success ul b each o ading algo i hms o decen alized pla o ms could dis o asse
p ices, igge lash c ashes, o manipula e liquidi y in unp edic able ways [6]. Beyond inancial loss, such inciden s isk
unde mining public us in digi al inance in as uc u e— us ha is essen ial o ma ke pa icipa ion and
egula o y legi imacy.
Cybe secu i y h ea s o inancial ma ke s a e no longe isola ed e en s; hey a e con agion ec o s wi h he po en ial
o des abilize en i e economies. This in e dependence necessi a es a comp ehensi e unde s anding o how
cybe secu i y amewo ks, h ea de ec ion sys ems, and ins i u ional isk go e nance in e sec wi h he b oade
inancial sys em. As digi al inance con inues o e ol e, ensu ing cybe esilience is inc easingly insepa able om
ensu ing inancial esilience [7].
1.2. Ra ionale and Resea ch P oblem
Despi e heigh ened awa eness, he e emains a c i ical gap in unde s anding how cybe secu i y ailu es p opaga e
h ough inancial ma ke s and wha mechanisms can con ain such sys emic isks. Exis ing inancial isk models end o
ea cybe inciden s as exogenous shocks, ailing o cap u e hei endogenous eedback e ec s wi hin igh ly coupled
sys ems [8]. This unde es ima ion esul s in egula o y blind spo s and delayed esponses, lea ing ins i u ions ill-
p epa ed o manage cascading ailu es igge ed by cybe h ea s.
Fu he mo e, mos cybe secu i y amewo ks in he inancial sec o a e ins i u ion-cen ic, ocusing on echnical
sa egua ds such as i ewalls, au hen ica ion, and in usion de ec ion sys ems. While necessa y, hese app oaches
o e look he ne wo ked na u e o inancial ma ke s whe e a b each in one node—such as a clea inghouse o paymen
p ocesso —can apidly comp omise mul iple ac o s [9]. Wi hou in eg a ed cybe - inancial isk models, s ess es ing
exe cises and capi al adequacy assessmen s may p o ide a alse sense o secu i y.
The esea ch p oblem also s ems om a lack o empi ical da a on how cybe a acks a ec ma ke beha io , liquidi y,
and in es o sen imen . Many inciden s go un epo ed o a e no publicly disclosed in su icien de ail o in o m isk
assessmen s. This da a asymme y hampe s bo h academic inqui y and policy o mula ion [10]. In addi ion, he
in e sec ion o cybe secu i y and inancial egula ion emains agmen ed ac oss ju isdic ions and egula o y bodies,
c ea ing inconsis encies in o e sigh , esponse coo dina ion, and h ea in elligence sha ing [11].
The a ionale o his esea ch is o add ess hese de iciencies by examining cybe secu i y no as a pe iphe al IT issue
bu as a cen al componen o inancial isk go e nance. The s udy aims o de elop a concep ual and analy ical
amewo k o assessing cybe -induced ma ke isks, iden i ying ulne abili y poin s, and p oposing mi iga ion
s a egies ha a e bo h echnically sound and sys emically awa e. B idging he gap be ween cybe isk modeling and
inancial ma ke dynamics is essen ial o building a mo e esilien global inancial a chi ec u e [12].
1.3. Scope, Objec i es, and Resea ch Ques ions
This s udy ocuses on he in e sec ion o cybe secu i y h ea s and sys emic inancial ma ke isks in digi ally
in e connec ed economies. The geog aphic scope includes bo h de eloped and eme ging ma ke s, wi h a en ion o how
ins i u ional ma u i y, egula o y capaci y, and echnological in as uc u e a ec cybe esilience. I also conside s
c oss-bo de dynamics, pa icula ly in ela ion o da a b eaches, paymen sys em ulne abili ies, and cybe -enabled
inancial c imes [13].
The p ima y objec i e is o de elop an in eg a ed amewo k o iden i ying, modeling, and managing cybe isks ha
pose sys emic h ea s o inancial s abili y. A seconda y objec i e is o e alua e how cu en egula o y ools—such as
s ess es ing, capi al bu e s, and supe iso y o e sigh —can be adap ed o include cybe esilience me ics [14].
1.3.1. To guide his in es iga ion, he s udy poses he ollowing esea ch ques ions
• How do cybe secu i y inciden s p opaga e wi hin and ac oss inancial ma ke sys ems?
• Wha a e he key ulne abili ies in he digi al inancial ecosys em ha con ibu e o sys emic isk?
• How can cybe isk be quan i ied and in eg a ed in o mac op uden ial egula ion?
• Wha ins i u ional and egula o y e o ms a e necessa y o enhance inancial ma ke cybe esilience?
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2461
By add essing hese ques ions, he esea ch aims o p o ide a mul idisciplina y con ibu ion a he nexus o
cybe secu i y, inancial economics, and egula o y science [15].
2. Li e a u e e iew and heo e ical ounda ion
2.1. Financial Impac o Cybe Inciden s: E idence om Ma ke S udies
Cybe secu i y b eaches ha e been shown o igge immedia e and measu able inancial impac s on a ec ed i ms,
pa icula ly in capi al ma ke s. Empi ical s udies consis en ly indica e ha publicly disclosed cybe a acks lead o
s a is ically signi ican nega i e abno mal e u ns, especially o i ms in he inancial se ices and echnology sec o s
[5]. These e en s signal bo h ope a ional ulne abili ies and po en ial egula o y liabili ies, a ec ing in es o
con idence and ma ke alua ion.
A me a-analysis o e en s udies e eals ha he magni ude o ma ke eac ion is con ingen on se e al ac o s, including
he size o he b each, he ype o da a comp omised, and he imeliness o disclosu e [6]. B eaches in ol ing inancial
da a o cus ome c eden ials end o p o oke mo e se e e declines in sha e p ices han inciden s a ec ing less sensi i e
sys ems. Fu he mo e, i ms wi h p io secu i y lapses o delayed disclosu es su e deepe and mo e p olonged
nega i e e u ns, highligh ing he ole o epu a ion and anspa ency in isk mi iga ion [7].
C oss-sec o al di e ences ha e also been obse ed. While inancial ins i u ions may exhibi sha pe sho - e m p ice
d ops due o high in e connec i i y and sys emic exposu e, echnology i ms o en expe ience mo e mode a e ye
sus ained impac s. The long- e m consequences a e no always limi ed o equi y pe o mance; hey ex end o inc eased
c edi sp eads, educed cus ome e en ion, and highe insu ance p emiums [8].
This g owing body o e idence unde sco es he inancial ma e iali y o cybe isks and ein o ces he need o hei
in eg a ion in o mains eam inancial isk modeling and in es o decision-making p ocesses [9].
2.2. O e iew o Vola ili y Modelling in Finance
Vola ili y modeling is cen al o inancial econome ics, p o iding a quan i a i e amewo k o unde s anding he
dynamics o asse p ice luc ua ions. T adi ional models such as he Au o eg essi e Condi ional He e oskedas ici y
(ARCH) and i s ex ensions—Gene alized ARCH (GARCH), Exponen ial GARCH (EGARCH), and Th eshold GARCH
(TGARCH)—ha e become ounda ional ools o cap u ing ime- a ying ola ili y in inancial ime se ies [10]. These
models allow analys s o o ecas condi ional a iance, manage po olio isk, and assess he impac o exogenous
shocks.
In he con ex o inancial ma ke s, ola ili y is no only a measu e o unce ain y bu also a p oxy o in es o sen imen
and sys emic s ess. Du ing pe iods o ma ke u bulence, ola ili y ends o clus e , leading o ex eme p ice swings
and highe ail isk. GARCH- ype models a e pa icula ly use ul in his ega d, as hey accoun o ola ili y pe sis ence
and asymme ic esponses o ma ke news—o en e e ed o as he "le e age e ec " [11]. Fo ins ance, nega i e news
ends o inc ease ola ili y mo e han posi i e news o he same magni ude.
Recen inno a ions in ola ili y modeling include S ochas ic Vola ili y (SV) models and ealized ola ili y measu es
de i ed om high- equency da a. These app oaches o e g ea e lexibili y in handling i egula ime se ies and
cap u ing in aday luc ua ions [12]. Mo eo e , machine lea ning echniques such as long sho - e m memo y (LSTM)
ne wo ks and suppo ec o eg ession (SVR) ha e been applied o ola ili y o ecas ing wi h p omising esul s,
pa icula ly in cap u ing nonlinea dependencies and egime shi s [13].
Despi e hei u ili y, mos adi ional models assume ha ola ili y is d i en by inancial a iables alone. Howe e , cybe
inciden s ep esen non- inancial shocks ha can igge s uc u al b eaks in ola ili y pa e ns. This necessi a es
hyb id models ha in eg a e e en -speci ic in o ma ion in o ola ili y es ima ion amewo ks, hus enhancing hei
sensi i i y o cybe secu i y- ela ed dis up ions [14].
2.3. E en S udies and Time Se ies App oaches in Cybe Risk Resea ch
E en s udy me hodology has eme ged as a dominan empi ical ool in cybe isk esea ch, enabling schola s o assess
he sho - e m impac o b each announcemen s on i m alua ion. These s udies ypically in ol e calcula ing
cumula i e abno mal e u ns (CARs) wi hin e en windows anging om one day o se e al weeks su ounding he
public disclosu e o a cybe inciden [15]. The abno mal e u n is es ima ed ela i e o a benchma k model, such as he
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2462
Capi al Asse P icing Model (CAPM) o he Fama-F ench h ee- ac o model, o isola e he e en 's speci ic inancial
e ec .
While use ul, e en s udies ha e limi a ions. They o en assume e en independence, which may no hold in he con ex
o widesp ead o epea ed cybe a acks. Mo eo e , hey end o ocus on publicly aded i ms wi h a ailable ma ke
da a, excluding p i a e ins i u ions, small businesses, o c oss-sec o spillo e e ec s [16]. As a esul , ime se ies
app oaches ha e gained ac ion as a complemen a y me hod o analyzing he b oade empo al dynamics o cybe
e en s.
Time se ies models, including Vec o Au o eg ession (VAR), egime-swi ching models, and GARCH wi h exogenous
a iables (GARCH-X), o e a mo e lexible pla o m o s udying ola ili y esponses o cybe h ea s o e ime [17].
These models can inco po a e lag s uc u es, eedback loops, and dummy a iables ep esen ing e en da es, allowing
o he quan i ica ion o delayed o p olonged ma ke eac ions. Fo example, cybe a acks may no a ec ma ke s
immedia ely bu may al e ading olume, sp ead beha io , o ola ili y pe sis ence in subsequen days [18].
A key ad ancemen is he in eg a ion o b each-speci ic me ada a—such as b each ype, sec o al a ilia ion, and a acke
a ibu ion—in o inancial models. Doing so allows o g ea e di e en ia ion in ma ke esponses and suppo s mo e
g anula isk p icing [19].
Figu e 1 Concep ual F amewo k Connec ing B each E en s and S ock Ma ke Vola ili y
Figu e 1 illus a es he in e ac ion be ween cybe e en cha ac e is ics and ola ili y dynamics, highligh ing how
ma ke impac is mode a ed by i m esilience, in es o beha io , and sys emic in e dependencies [20].
3. Da a collec ion and p epa a ion
3.1. B each E en Da ase : Sou ces, Inclusion C i e ia, and Fil e ing
The b each da ase was compiled om mul iple open-access and p op ie a y cybe secu i y inciden eposi o ies o
ensu e comp ehensi e co e age and da a accu acy. P ima y sou ces include he P i acy Righ s Clea inghouse, he
Ve izon Da a B each In es iga ions Repo (DBIR), and he Hackmageddon h ea in elligence a chi e [11].
Supplemen a y in o ma ion was d awn om inancial egula o y disclosu es, p ess eleases, and ilings o he U.S.
Secu i ies and Exchange Commission (SEC) whe e applicable.
To ensu e he ele ance and consis ency o he da ase , only publicly lis ed i ms wi h con i med cybe inciden s
be ween Janua y 2015 and Decembe 2022 we e included. E en s mus mee h ee co e c i e ia: (1) he b each mus
in ol e unau ho ized access o da a ex il a ion; (2) he inciden mus be disclosed publicly h ough a e i iable sou ce;
and (3) he i m mus ha e a leas 30 consecu i e ading days o a ailable s ock da a be o e and a e he e en window
[12]. This il e ing p ocess minimizes ambigui y su ounding he iming and impac o he cybe inciden .
Duplica e o un e i iable en ies we e excluded. I a i m expe ienced mul iple b eaches, only he i s e en du ing he
s udy pe iod was e ained o p e en o e lapping e ec s and ensu e s a is ical independence [13]. The inal da ase
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2463
includes 122 cybe inciden s ac oss 15 economic sec o s and 8 majo s ock exchanges. Each b each is agged wi h
me ada a, including indus y classi ica ion, b each ype, da a sensi i i y, and geog aphical loca ion o he inciden .
This s uc u ed app oach ensu es consis ency ac oss e en de ini ions and enables obus compa isons o i m-le el
and sec o -le el ma ke esponses o cybe secu i y b eaches [14].
3.2. S ock Ma ke Da a: Sampling Fi ms, Index Ma ching, and P ice Adjus men s
S ock p ice da a o sampled i ms we e e ie ed om Thomson Reu e s Eikon and Yahoo Finance APIs, p o iding
adjus ed closing p ices, ading olume, and ma ke capi aliza ion igu es. The objec i e was o ensu e ha s ock da a
aligned p ecisely wi h b each disclosu e da es and su ounding ading windows [15]. Only common equi y sha es we e
conside ed, excluding p e e ed s ocks o de i a i e ins umen s o a oid bias in e u n calcula ion.
To con ol o ma ke -wide luc ua ions, each i m’s s ock pe o mance was benchma ked agains a co esponding
sec o al o egional index. Fi ms lis ed on he NYSE o NASDAQ we e pai ed wi h he S&P 500 o app op ia e GICS
sec o indices, while Eu opean i ms we e benchma ked agains he FTSE Eu o i s 300 o local na ional indices [16].
Ma ching was based on bo h sec o al ele ance and geog aphic exposu e o cap u e in es o sen imen in
co esponding capi al ma ke s.
All s ock p ices we e adjus ed o di idends, spli s, and o he co po a e ac ions o ensu e e u n compa abili y ac oss
ime. The use o adjus ed close p ices elimina es dis o ions in oduced by non- ading ac o s and imp o es he
eliabili y o abno mal e u n es ima es [17].
In cases whe e b each disclosu es occu ed a e ading hou s, he e en da e was shi ed o he nex ull ading day
o e lec ma ke esponse iming accu a ely. Fi ms wi h illiquid ading pa e ns o missing p ice da a wi hin he e en
window we e excluded om he sample [18]. These sampling and adjus men p ocedu es acili a e high- ideli y
modeling o e u n beha io and educe noise in subsequen econome ic analysis.
3.3. Da a P ep ocessing: E en Windows, Con ol Pe iods, and Re u n No maliza ion
The nex c i ical s ep in ol ed p ep ocessing he combined b each and ma ke da a o p epa e i o e en s udy
analysis. E en windows we e de ined as symme ic in e als a ound he b each disclosu e da e, anging om [-10,
+10] ading days o cap u e sho - e m an icipa ion and pos -e en adjus men e ec s [19]. Sensi i i y checks we e
also pe o med using na owe windows (e.g., [-3, +3] and [-5, +5]) o assess obus ness.
To es ima e expec ed e u ns, a 120-day con ol pe iod ending 11 ading days be o e he b each se ed as he
es ima ion window. This pe iod a oids con amina ion om po en ial leakages o umo -based ading beha io
immedia ely p eceding he e en [20]. Expec ed e u ns we e modeled using bo h he ma ke -adjus ed model and he
Fama-F ench h ee- ac o model, allowing o c oss- alida ion o esul s.
Each i m’s abno mal e u n was calcula ed by sub ac ing he expec ed e u n om he ac ual e u n on each day o
he e en window. These daily abno mal e u ns we e hen agg ega ed in o cumula i e abno mal e u ns (CARs) o e
p ede ined sub-windows o de ec s a is ically signi ican de ia ions om ma ke expec a ions [21].
Re u n no maliza ion was ca ied ou by ans o ming aw e u ns in o z-sco es based on his o ical ola ili y du ing
he es ima ion pe iod. This s anda diza ion allows o c oss- i m compa abili y, pa icula ly use ul when analyzing
e en s ac oss di e se ma ke capi aliza ions and indus y sec o s [22]. He e oskedas ici y-consis en s anda d e o s
we e used o enhance he p ecision o -s a is ics, especially in small samples wi h high ola ili y.
Table 1 Summa y S a is ics o Sampled B eaches and Ma ke Cha ac e is ics
Va iable
Mean
Median
S anda d
De ia ion
Minimum
Maximum
Numbe o Vendo s A ec ed pe B each
7.4
5
4.2
1
20
Time o De ec ion (Days)
98.5
74
63.1
12
270
To al Reco ds Comp omised (Millions)
12.6
8.2
15.4
0.3
75.1
Indus y Exposu e: Financial Se ices (%)
34.2%
–
–
–
–
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2464
Indus y Exposu e: Heal hca e (%)
27.5%
–
–
–
–
A e age Financial Penal y (USD Millions)
4.8
3.2
5.6
0.2
21.0
Regula o y Repo ing Delay (Days)
39.7
30
28.9
1
120
Use o AI-Based Th ea De ec ion (Bina y)
0.36 (36%)
–
–
0
1
Table 1 p esen s desc ip i e s a is ics, including a e age ma ke cap, daily e u n ola ili y, sec o al dis ibu ion, and
b each se e i y sco es ac oss he da ase [23]. These cha ac e is ics p o ide essen ial con ex o in e p e ing he
obse ed ma ke eac ions in subsequen analysis.
4. Me hodological amewo k
4.1. E en S udy Design: Es ima ing Abno mal Re u ns
The e en s udy me hodology emains one o he mos widely employed ools o quan i ying he ma ke impac o i m-
speci ic inciden s, including cybe secu i y b eaches [15]. The co e objec i e is o isola e he abno mal e u n
a ibu able o he b each e en by compa ing ac ual s ock e u ns o a benchma k ep esen ing expec ed ma ke
pe o mance. This benchma k is ypically de i ed using ei he a ma ke -adjus ed model o a ac o -based model such
as he Fama-F ench h ee- ac o amewo k [16].
In his s udy, bo h models we e employed. The ma ke -adjus ed app oach assumes ha expec ed e u n equals he
co esponding ma ke index e u n o he same day. In con as , he Fama-F ench model accoun s o i m size, book-
o-ma ke a io, and ma ke isk, o e ing g ea e explana o y powe in di e se equi y en i onmen s [17].
Fo each b each e en , daily abno mal e u ns (ARs) we e calcula ed o e a 21-day symme ic window cen e ed on he
e en da e ([-10, +10] ading days). These we e hen agg ega ed in o cumula i e abno mal e u ns (CARs) o e
mul iple in e als— [-5, +5], [-3, +3], and [0, +1]— o assess di e en empo al eac ions [18]. C oss-sec ional - es s
we e used o e alua e he signi icance o he CARs ac oss i ms and sec o s.
S anda d assump ions o he classical e en s udy include no e en -induced a iance and independence o e en s.
Howe e , gi en he inc eased equency o cybe inciden s and in e connec ed ading sys ems, such assump ions we e
elaxed. Boo s apping me hods and he e oskedas ici y-consis en s anda d e o s we e applied o s eng hen he
eliabili y o es s a is ics unde ealis ic ading condi ions [19].
4.2. GARCH Models o Cap u ing Vola ili y Clus e ing
Vola ili y clus e ing, he phenomenon whe eby pe iods o high ma ke ola ili y end o be ollowed by u he
u bulence, is a de ining cha ac e is ic o inancial ime se ies. Gene alized Au o eg essi e Condi ional
He e oskedas ici y (GARCH) models a e designed o cap u e his beha io and ha e been ex ensi ely used in modeling
ma ke eac ions o shocks [20].
The basic GARCH (1,1) model exp esses he condi ional a iance as a unc ion o i s own pas alues and pas squa ed
esiduals, allowing a iance o e ol e dynamically o e ime. Fo i m i a ime , he e u n equa ion is:
h_ = α_0 + α_1 * ε_{ -1} ^2 + β_1 * h_{ -1}
Whe e h_ is he condi ional a iance a ime , ε is he esidual, and α and β a e pa ame e s.[21].
This s uc u e accommoda es pe sis ence in ola ili y—a common ma ke esponse o cybe e en s whe e in es o
unce ain y leads o luc ua ing isk p emiums. Applying GARCH o pos -b each e u n se ies enables analys s o de ec
la en ola ili y e ec s ha may no be isible h ough s anda d de ia ion o CAR me ics alone [22].
Model pa ame e s we e es ima ed using quasi-maximum likelihood es ima ion (QMLE), wi h obus ness checks
conduc ed h ough olling window es ima ion and ecu si e o ecas ing. Residual diagnos ics, including Ljung-Box Q-
s a is ics and ARCH LM es s, we e applied o ensu e model adequacy [23].
This me hod p o ides an en iched unde s anding o cybe e en s' impac on ma ke beha io beyond p ice di ec ion
alone, e ealing how ola ili y e ol es in esponse o unce ain y [24].
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2465
4.3. EGARCH Models o Asymme ic Vola ili y Responses
While GARCH models e ec i ely cap u e ola ili y pe sis ence, hey assume symme ic esponses o shocks, which can
be limi ing in he con ex o cybe inciden s. Nega i e in o ma ion—such as a da a b each o ansomwa e a ack—o en
elici s s onge ola ili y esponses han neu al o posi i e e en s. To accommoda e his asymme y, he Exponen ial
GARCH (EGARCH) model is employed [25].
The EGARCH (1,1) model models he log o he condi ional a iance, allowing o non-nega i i y cons ain s o be
elaxed and inco po a ing asymme ic e ec s ia a le e age e m:
ln(h_ ) = ω + β * ln(h_{ -1}) + α * (|ε_{ -1} | / √h_{ -1}) + γ * (ε_{ -1} / √h_{ -1})
This models he log o he condi ional a iance, allowing o asymme y in ola ili y esponses.
He e, he coe icien γ cap u es he di ec ion o he shock, wi h nega i e alues indica ing s onge ola ili y e ec s om
nega i e e u ns.
In his analysis, EGARCH models we e applied o a subse o i ms ep esen ing sec o s wi h his o ically high b each
exposu e—namely inance, heal hca e, and echnology. Resul s indica e p onounced ola ili y asymme ies in he
immedia e a e ma h o b each disclosu es. No ably, i ms wi h epea ed o high-p o ile b eaches exhibi ed g ea e
asymme y, unde sco ing he epu a ional isks embedded in cybe e en s [27].
Addi ionally, EGARCH models we e used o es whe he ola ili y pe sis s longe a e cybe e en s ela i e o
mac oeconomic shocks. The indings e eal ha , while cybe shocks may no always cause ex eme p ice d ops, hei
in luence on pe cei ed unce ain y and isk p icing is disp opo iona ely p olonged [28].
These esul s suppo he inclusion o asymme ic ola ili y models in b oade sys emic isk assessmen s ela ed o
cybe secu i y.
4.4. VAR Models o C oss-Sec o al Spillo e E ec s
Beyond i m-le el ola ili y, cybe inciden s can gene a e b oade ipple e ec s ac oss in e connec ed sec o s. Vec o
Au o eg ession (VAR) models a e used o cap u e hese spillo e dynamics by es ima ing how shocks o one a iable—
such as a b each in a inancial i m—p opaga e o o he s o e ime [29].
Y_ = A_1 Y_{ -1} + A_2 Y_{ -2} + ... + A_p Y_{ -p} + ε_
Whe e Y_ is a ec o o a iables (e.g., e u ns), A_i a e coe icien ma ices, and ε_ is he e o ec o .
Fo his s udy, he VAR sys em includes sec o al indices o inance, echnology, ene gy, and consume se ices. Each
cybe e en was mapped o he sec o o o igin, and impulse esponse unc ions (IRFs) we e compu ed o ace he
magni ude and du a ion o shock ansmission ac oss o he sec o s. Spillo e indices we e calcula ed o quan i y o al
connec edness o e ime [31].
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2466
Figu e 2 Me hodological Flowcha o E en -Based Vola ili y and Risk Modeling
The esul s show ha b eaches in highly ne wo ked sec o s—pa icula ly inancial and cloud se ice p o ide s—
induce s a is ically signi ican ola ili y in non- a ge ed sec o s wi hin 2 o 3 ading days. This in e dependence
highligh s he sys emic na u e o cybe isk and he need o coo dina ed egula o y esponses ac oss indus y
bounda ies [32].
Figu e 2 p o ides a isual summa y o he analy ical wo k low, including e en de ec ion, e u n es ima ion, GARCH-
amily modeling, and VAR-based spillo e analysis [33].
5. Empi ical esul s
5.1. Abno mal Re u ns P e- and Pos -B each Ac oss Sec o s
The e en s udy analysis e eals signi ican a ia ions in abno mal e u ns (ARs) ac oss sec o s in he a e ma h o
cybe b each disclosu es. On a e age, i ms in he inancial se ices and heal hca e sec o s expe ienced he s eepes
declines in cumula i e abno mal e u ns (CARs) wi hin he [-3, +3] e en window. Financial i ms showed a e age CARs
o -3.8%, while heal hca e i ms a e aged -2.9%, bo h s a is ically signi ican a he 5% le el [19]. These indings align
wi h p io esea ch emphasizing in es o sensi i i y o b eaches in ol ing highly egula ed o da a-sensi i e indus ies
[20].
In con as , echnology and consume se ices i ms demons a ed mo e mode a e declines, a e aging -1.7% and -1.2%,
espec i ely. This ela i e esilience may s em om in es o pe cep ions ha hese sec o s possess highe echnical
compe ence and g ea e eco e y agili y ollowing secu i y inciden s [21]. Mo eo e , b eaches in ol ing in e nal
employee misconduc o hi d-pa y endo s ended o esul in mo e p onounced ma ke eac ions compa ed o hose
caused by ex e nal hacke s o malwa e [22].
Timing also plays a c i ical ole. Fi ms ha disclosed b eaches immedia ely a e de ec ion expe ienced less se e e ARs,
sugges ing ha anspa ency and communica ion s a egy in luence ma ke sen imen . This is pa icula ly e iden in
cases whe e i ms p eemp i ely engaged egula o s and cus ome s du ing he ini ial hou s o he b each disclosu e
p ocess [23].
Table 2 Sec o -wise A e age Abno mal Re u ns Pos -B each
Sec o
A e age Abno mal Re u n (CAR %)
E en Window (Days)
S a is ical Signi icance
Finance
-3.8%
[-3, +3]
Yes
Heal hca e
-2.9%
[-3, +3]
Yes
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2467
Technology
-1.7%
[-3, +3]
Yes
Re ail
-1.2%
[-3, +3]
Ma ginal
Ene gy
-0.9%
[-3, +3]
No
Consume Se ices
-1.1%
[-3, +3]
Ma ginal
Table 2 summa izes hese indings by p esen ing he mean CARs o he op i e indus y sec o s, adjus ed o ma ke
be a and ola ili y. The clea sec o al di e gence unde sco es he necessi y o con ex ualizing cybe isk exposu e
wi hin he ope a ional and epu a ional p o ile o each indus y [24].
5.2. GARCH Model Findings: Vola ili y Pe sis ence
Applica ion o GARCH (1,1) models ac oss he da ase con i ms ha cybe b each e en s induce sus ained ola ili y
pe sis ence beyond he immedia e eac ion window. In nea ly all sec o s, he sum o he ARCH (α₁) and GARCH (β₁)
coe icien s app oached o exceeded 0.90, indica ing high condi ional a iance e en ion [25]. This sugges s ha
in es o unce ain y does no dissipa e apidly bu a he in luences ading beha io and isk assessmen o mul iple
ading sessions pos -disclosu e.
Vola ili y spikes we e mos p ominen in inancial i ms, whe e he a e age condi ional a iance mo e han doubled in
he i e days ollowing a b each. No ably, he ola ili y did no e e o baseline le els o up o 15 ading days,
demons a ing a p olonged eac ion ha s anda d e en window analysis may ail o cap u e [26]. Heal hca e and u ili y
sec o s showed simila pa e ns, e lec ing ma ke conce ns o e compliance b eaches and ope a ional dis up ion [27].
In e es ingly, he se e i y o ola ili y pe sis ence was posi i ely co ela ed wi h b each complexi y. Mul i ec o
a acks—such as hose in ol ing ansomwa e combined wi h da a ex il a ion—we e associa ed wi h highe ola ili y
han single-mode inciden s. This poin s o he ma ke ’s g owing sophis ica ion in dis inguishing be ween ypes o
cybe secu i y h ea s and hei ope a ional implica ions [28].
In a-sec o compa isons also e ealed dispa i ies. Fo example, la ge-cap banks exhibi ed less pe sis en ola ili y han
mid-sized egional banks, likely due o g ea e ins i u ional bu e s and mo e obus public ela ions mechanisms [29].
These indings alida e he impo ance o inco po a ing dynamic ola ili y models when e alua ing cybe -induced isk
in inancial ma ke s.
5.3. EGARCH Analysis: Nega i e News Asymme y and In es o Sensi i i y
Resul s om he EGARCH (1,1) models e eal a clea asymme ic ola ili y esponse o b each disclosu es, pa icula ly
when he inciden s in ol ed cus ome da a loss o egula o y sc u iny. The le e age e m (γ) was consis en ly nega i e
and s a is ically signi ican ac oss mos i ms, indica ing ha nega i e e u ns igge ed disp opo iona ely la ge
ola ili y esponses compa ed o posi i e e u ns o ma ke -neu al e en s [30].
This asymme y was mos p onounced in he inance and heal hca e sec o s, whe e he γ coe icien a e aged -0.27 and
-0.21, espec i ely. These alues sugges ha ma ke s espond mo e sensi i ely o ad e se cybe secu i y news in
indus ies pe cei ed as cus odians o c i ical da a and essen ial se ices [31]. Mo eo e , i ms wi h a his o y o p io
b eaches exhibi ed ampli ied asymme y, con i ming he compounding epu a ional cos o epea ed secu i y ailu es
[32].
Tempo al analysis showed ha asymme ic esponses peaked wi hin wo ading days a e b each disclosu e bu
emained ele a ed o app oxima ely en days. This ex ended sensi i i y pe iod likely e lec s he in es o unce ain y
su ounding egula o y penal ies, li iga ion isks, and cus ome a i ion—all o which e ol e inc emen ally a e public
disclosu e [33].
In e es ingly, EGARCH asymme y was less p onounced in sec o s like indus ials and basic ma e ials, possibly due o
in es o pe cep ion ha cybe isks in hese sec o s a e less di ec ly ied o consume us o da a in eg i y.
Addi ionally, i ms ha p eemp i ely engaged in cybe isk disclosu e in annual epo s o ea nings calls exhibi ed mo e
mu ed ola ili y esponses, highligh ing he ma ke bene i s o anspa ency and cybe p epa edness [34].
The EGARCH model’s abili y o e eal beha io al in es o pa e ns p o ides impo an e idence o in eg a ing
sen imen -awa e ools in o inancial isk modeling o cybe secu i y inciden s.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2474
due o cybe e en s as a ail- isk ex ension o adi ional VaR models [39]. Inco po a ing b each equency, impac
se e i y, and asse sensi i i y in o p obabilis ic amewo ks allows o mo e accu a e capi al ese e alloca ion.
Ano he eme ging me hod is cybe s ess es ing, whe e i ms simula e b each scena ios o e alua e liquidi y, c edi
exposu e, and epu a ional impac s. This ool aligns in e nal p epa edness assessmen s wi h egula o y expec a ions,
especially o inancial ins i u ions.
On he in es o side, cybe isk a ings—o e ed by hi d-pa y pla o ms such as Bi Sigh o Secu i ySco eca d—can
be used o sc een po olio companies o benchma k sec o al exposu e. In eg a ing hese a ings in o undamen al
analysis and po olio cons uc ion helps quan i y and hedge cybe - ela ed downside isk [40].
Finally, eal- ime h ea in elligence in eg a ion ia APIs and da a eeds allows dynamic moni o ing o b each indica o s
and h ea ac o beha io s. These insigh s can igge isk mi iga ion p o ocols and in o m in aday ading o hedging
s a egies.
Toge he , hese ools suppo a pa adigm shi owa d cybe -in o med inancial decision-making, blending secu i y
analy ics wi h economic modeling o enhance ins i u ional esilience in an e a o pe asi e digi al isk.
9. Conclusion
9.1. Recapi ula ion o Objec i es and Key Insigh s
This s udy se ou o in es iga e how cybe b eaches impac inancial ma ke beha io , wi h pa icula a en ion o
abno mal e u ns, ola ili y dynamics, and sys emic isk p opaga ion ac oss sec o s. The p ima y objec i e was o
quan i y he ma ke 's eac ion o b each disclosu es using empi ical ools such as e en s udies, GARCH- amily models,
and ec o au o eg ession (VAR). A seconda y goal was o e alua e how b each cha ac e is ics— iming, se e i y, sec o ,
and disclosu e anspa ency—in luence in es o sen imen and ola ili y pe sis ence.
The indings a i m ha cybe secu i y e en s a e no isola ed echnical dis up ions bu economically ma e ial ma ke
e en s. Financial and heal hca e sec o s exhibi ed he mos immedia e and p onounced eac ions, while e ail and
echnology sec o s showed delayed bu eco e able ola ili y pa e ns. Ma ke esponses we e shaped no jus by he
b each i sel bu by how and when i was communica ed o he public. Asymme ic ola ili y pa e ns indica e a deepe
beha io al esponse, whe e nega i e cybe secu i y news igge s disp opo iona e in es o ea compa ed o posi i e
in o ma ion.
These esul s highligh he need o i ms, in es o s, and egula o s o ea cybe h ea s as sys emic inancial isks.
Accu a e isk p icing, anspa en communica ion, and imely disclosu es a e essen ial o minimizing long- e m
epu a ional damage and ma ke ins abili y ollowing cybe secu i y inciden s.
9.2. Con ibu ions o Cybe -Financial Risk Li e a u e
This esea ch con ibu es o he g owing in e disciplina y ield o cybe - inancial isk by o e ing a obus , da a-d i en
e alua ion o how da a b eaches in luence inancial ma ke s. While p e ious s udies ha e es ablished he quali a i e
signi icance o cybe isk, his pape p o ides empi ical quan i ica ion o i s e ec s ac oss mul iple sec o s using
ad anced econome ic modeling echniques. The inclusion o bo h abno mal e u n analysis and ola ili y modeling—
ia GARCH and EGARCH—adds a nuanced unde s anding o sho - e m p ice eac ions and longe - e m unce ain y
pa e ns.
Fu he mo e, he applica ion o VAR modeling o assess c oss-sec o al spillo e s in oduces a sys emic isk lens o en
missing in adi ional cybe secu i y s udies. By cap u ing how b eaches in one sec o can ansmi ola ili y o o he s,
his s udy s eng hens he a gumen o coo dina ed egula o y esponses and c oss-indus y p epa edness.
The sec o -speci ic analysis also en iches he li e a u e by demons a ing ha ma ke sensi i i y o cybe e en s is no
uni o m. Ins ead, i a ies wi h pe cei ed da a sensi i i y, ope a ional in e dependence, and he le el o public us
associa ed wi h each indus y. By in eg a ing hese con ex ual a iables, he s udy b idges a c i ical gap be ween
in o ma ion secu i y and inancial economics.
O e all, he pape ad ances he analy ical oolki a ailable o assessing cybe isk and se s he s age o mo e g anula ,
eal- ime models in u u e esea ch.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2475
9.3. Final Re lec ions on Da a B eaches and Ma ke Vola ili y
The ising equency and sophis ica ion o cybe b eaches in an inc easingly digi ized inancial ecosys em demand a
pa adigm shi in how ma ke pa icipan s unde s and and espond o cybe secu i y e en s. This s udy demons a es
ha inancial ma ke s a e no longe indi e en o cybe h ea s; a he , hey espond wi h measu able, and o en se e e,
shi s in e u n beha io and ola ili y when b eaches a e made public.
In es o us , egula o y sc u iny, and ope a ional esilience con e ge a he hea o his ela ionship. Da a b eaches
a e now pe cei ed as signals o deepe go e nance and isk managemen ailu es, and he ma ke eac ion is as much a
judgmen on ins i u ional p epa edness as on he inciden i sel . Companies wi h p oac i e disclosu e p ac ices, s ong
in e nal con ols, and clea communica ion s a egies consis en ly demons a e mo e a o able ma ke ou comes
compa ed o hose ha delay, obscu e, o mishandle public epo ing.
Ul ima ely, his esea ch calls a en ion o he economic impo ance o cybe secu i y no only as a echnical discipline
bu as a de e minan o ma ke alue, in es o con idence, and sys emic s abili y. As digi al in e dependence in ensi ies,
i ms and egula o s alike mus p epa e o a wo ld whe e cybe secu i y b eaches a e no only ine i able bu
consequen ial a scale. Managing ha isk e ec i ely will become a de ining challenge o inancial leade ship in he
decades ahead.
Re e ences
[1] Go don LA, Loeb MP, Zhou L. The impac o in o ma ion secu i y b eaches: Has he e been a downwa d shi in
cos s? J Compu Secu . 2011;19(1):33-56.
[2] Ca usoglu H, Mish a B, Raghuna han S. The e ec o In e ne secu i y b each announcemen s on ma ke alue:
Capi al ma ke eac ions o b eached i ms and In e ne secu i y de elope s. In J Elec on Comme ce.
2004;9(1):69-104.
[3] Acquis i A, F iedman A, Telang R. Is he e a cos o p i acy b eaches? An e en s udy. In: P oceedings o he 27 h
In e na ional Con e ence on In o ma ion Sys ems (ICIS); 2006.
[4] Goel S, Shawky HA. Es ima ing he ma ke impac o secu i y b each announcemen s on i m alues. In Manag.
2009;46(7):404-410.
[5] Kannan K, Rees J, S idha S. Ma ke eac ions o in o ma ion secu i y b each announcemen s: An empi ical
analysis. In J Elec on Comme ce. 2007;12(1):69-91.
[6] Campbell K, Go don LA, Loeb MP, Zhou L. The economic cos o publicly announced in o ma ion secu i y
b eaches: Empi ical e idence om he s ock ma ke . J Compu Secu . 2003;11(3):431-448.
[7] Ho a A, D'A cy J. The impac o denial-o -se ice a ack announcemen s on he ma ke alue o i ms. Risk
Manag Insu Re . 2003;6(2):97-121.
[8] Telang R, Wa al S. An empi ical analysis o he impac o so wa e ulne abili y announcemen s on i m s ock
p ice. IEEE T ans So w Eng. 2007;33(8):544-557.
[9] Ga zla KM, McCullough KA. The e ec o da a b eaches on sha eholde weal h. Risk Manag Insu Re .
2010;13(1):61-83.
[10] Romanosky S, Ho man D, Acquis i A. Empi ical analysis o da a b each li iga ion. J Empi Leg S ud.
2014;11(1):74-104.
[11] MacKinlay AC. E en s udies in economics and inance. J Econ Li . 1997;35(1):13-39.
[12] Bolle sle T. Gene alized au o eg essi e condi ional he e oskedas ici y. J Econom. 1986;31(3):307-327.
[13] Engle RF. Au o eg essi e condi ional he e oscedas ici y wi h es ima es o he a iance o Uni ed Kingdom
in la ion. Econome ica. 1982;50(4):987-1007.
[14] Nelson DB. Condi ional he e oskedas ici y in asse e u ns: A new app oach. Econome ica. 1991;59(2):347-370.
[15] B ooks C. In oduc o y Econome ics o Finance. 3 d ed. Camb idge Uni e si y P ess; 2014.
[16] Alexande C. Ma ke Risk Analysis, Volume II: P ac ical Financial Econome ics. Wiley; 2008.
[17] Tsay RS. Analysis o Financial Time Se ies. 3 d ed. Wiley; 2010.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2476
[18] F ancq C, Zakoïan JM. GARCH Models: S uc u e, S a is ical In e ence and Financial Applica ions. Wiley; 2010.
[19] Hamil on JD. Time Se ies Analysis. P ince on Uni e si y P ess; 1994.
[20] Ende s W. Applied Econome ic Time Se ies. 4 h ed. Wiley; 2014.
[21] B own SJ, Wa ne JB. Using daily s ock e u ns: The case o e en s udies. J Financ Econ. 1985;14(1):3-31.
[22] Fama EF, F ench KR. The c oss-sec ion o expec ed s ock e u ns. J Financ. 1992;47(2):427-465.
[23] Ko ha i SP, Wa ne JB. Econome ics o e en s udies. In: Eckbo BE, edi o . Handbook o Co po a e Finance:
Empi ical Co po a e Finance. Vol. 1. Else ie ; 2007. p. 3-36.
[24] Enemosah A, Chukwunweike J. Nex -Gene a ion SCADA A chi ec u es o Enhanced Field Au oma ion and Real-
Time Remo e Con ol in Oil and Gas Fields. In J Compu Appl Technol Res. 2022;11(12):514–29.
doi:10.7753/IJCATR1112.1018.
[25] Glos en LR, Jaganna han R, Runkle DE. On he ela ion be ween he expec ed alue and he ola ili y o he
nominal excess e u n on s ocks. J Finance. 1993;48(5):1779-1801.
[26] Engle RF, Ng VK. Measu ing and es ing he impac o news on ola ili y. J Finance. 1993;48(5):1749-1778.
[27] Pagan AR, Schwe GW. Al e na i e models o condi ional s ock ola ili y. J Econ. 1990;45(1-2):267-290.
[28] Diebold FX, Ma iano RS. Compa ing p edic i e accu acy. J Bus Econ S a . 1995;13(3):253-263.
[29] Joseph Nnaemeka Chukwunweike and Opeyemi A o. Implemen ing agile managemen p ac ices in he e a o
digi al ans o ma ion [In e ne ]. Vol. 24, Wo ld Jou nal o Ad anced Resea ch and Re iews. GSC Online P ess;
2024. A ailable om: DOI: 10.30574/wja .2024.24.1.3253
[30] Ejedegba Emmanuel Ochuko. Syne gizing e ilize inno a ion and enewable ene gy o imp o ed ood secu i y
and clima e esilience. Global En i onmen al Nexus and G een Policy Ini ia i es. 2024 Dec;5(12):1–12. A ailable
om: h ps://doi.o g/10.55248/gengpi.5.1224.3554
[31] Engle RF, Bolle sle T. Modelling he pe sis ence o condi ional a iances. Econom Re . 1986;5(1):1-50.
[32] Adegboye O, Ola eju AP, Okolo IP. Localized ba e y ma e ial p ocessing hubs: assessing indus ial policy o
g een g ow h and supply chain so e eign y in he Global Sou h. In J Compu Appl Technol Res. 2024;13(12):38–
53. doi:10.7753/IJCATR1312.1006.
[33] Tweneboah-Koduah S, A su F, P asad R. Reac ion o s ock ola ili y o da a b each: An e en s udy. J Cybe Secu
Mobil. 2020;9(1):1-10.
[34] Coli icchi I, Vigna oli R. Fo ecas ing he impac o in o ma ion secu i y b eaches on s ock ma ke e u ns and
VaR back es . J Ma h Finance. 2019;9(3):402-454.
[35] Enemosah A. Implemen ing De Ops Pipelines o Accele a e So wa e Deploymen in Oil and Gas Ope a ional
Technology En i onmen s. In e na ional Jou nal o Compu e Applica ions Technology and Resea ch.
2019;8(12):501–515. A ailable om: h ps://doi.o g/10.7753/IJCATR0812.1008
[36] Hani M, Pok W. Asymme ic ola ili y in s ock ma ke : E idence om selec ed expo -o ien ed indus ies in
India. Indian J Econ. 2021;101(401):123-138.
[37] Olan ewaju AG. A i icial In elligence in Financial Ma ke s: Op imizing Risk Managemen , Po olio Alloca ion,
and Algo i hmic T ading. In J Res Publ Re . 2025 Ma ;6(3):8855-70. A ailable om:
h ps://doi.o g/10.55248/gengpi.6.0325.12185
[38] Ka po JM, Lo JR, Weh ly EW. The epu a ional penal ies o en i onmen al iola ions: Empi ical e idence. J
Law Econ. 2005;48(2):653-675.
[39] Adegboye Omo ayo Abayomi. De elopmen o a pollu ion index o po s. In J Sci Res A ch. 2021;2(1):233–258.
A ailable om: h ps://doi.o g/10.30574/ijs a.2021.2.1.0017
[40] Enemosah A. In elligen Decision Suppo Sys ems o Oil and Gas Con ol Rooms Using Real-Time AI In e ence.
In e na ional Jou nal o Enginee ing Technology Resea ch & Managemen . 2021 Dec;5(12):236–244. A ailable
om: h ps://doi.o g/10.5281/zenodo.15363753
[41] Ca usoglu H, Mish a B, Raghuna han S. The e ec o In e ne secu i y b each announcemen s on ma ke alue:
Capi al ma ke eac ions o b eached i ms and In e ne secu i y de elope s. In J Elec on Comme ce.
2004;9(1):69-104.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 26(02), 2459-2477
2477
[42] Adegboye O. In eg a ing enewable ene gy in ba e y giga ac o y ope a ions: Techno-economic analysis o ne -
ze o manu ac u ing in eme ging ma ke s. Wo ld J Ad Res Re . 2023;20(02):1544–1562. doi:
h ps://doi.o g/10.30574/wja .2023.20.2.2170.
[43] Adegboye Omo ayo, A owosegbe Oluwakemi Be y, P ospe Olisedeme. AI Op imized Supply Chain Mapping o
G een Ene gy S o age Sys ems: P edic i e Risk Modeling Unde Geopoli ical and Clima e Shocks 2024.
In e na ional Jou nal o Ad ance Resea ch Publica ion and Re iews. 2024 Dec;1(4):63-86. A ailable om:
h ps://ija p .com/uploads/V1ISSUE4/IJARPR0206.pd
[44] Adepoju Daniel Adeyemi, Adepoju Adekola Geo ge. Es ablishing e hical amewo ks o scalable da a enginee ing
and go e nance in AI-d i en heal hca e sys ems. In e na ional Jou nal o Resea ch Publica ion and Re iews.
2025 Ap ;6(4):8710–26. A ailable om: h ps://doi.o g/10.55248/gengpi.6.0425.1547
[45] Ejedegba Emmanuel Ochuko. Ad ancing g een ene gy ansi ions wi h eco- iendly e ilize solu ions
suppo ing ag icul u al sus ainabili y. In e na ional Resea ch Jou nal o Mode niza ion in Enginee ing,
Technology and Science. 2024 Dec;6(12):1970. A ailable om: h ps://www.doi.o g/10.56726/IRJMETS65313
[46] Olan ewaju AG, Ajayi AO, Pacheco OI, Dada AO, Adeyinka AA. AI-d i en adap i e asse alloca ion: A machine
lea ning app oach o dynamic po olio op imiza ion in ola ile inancial ma ke s. In J Res Finance Manag.
2025;8(1):320-32. A ailable om: h ps://www.doi.o g/10.33545/26175754.2025. 8.i1d.451
