scieee Science in your language
[en] (orig)

Prediction of business cycle of Poland

Abstract

Tomas Bata University in Zlin, TBU: RVO/2022

Read accessible full text

Prediction of business cycle of Poland

Author: Tkáčová, Andrea,Gavurová, Beáta,Kelemen, Katarína
Publisher: Centre of Sociological Research
Year: 2022
DOI: 10.14254/2071-8330.2022/15-3/5
Source: https://publikace.k.utb.cz/bitstream/10563/1011203/1/Fulltext_1011203.pdf
65
Tkaco a, A., Ga u o a, B., & Kelemen, K. (2022). P edic ion o business cycle o
Poland. Jou nal o In e na ional S udies, 15(3), 65-81. doi:10.14254/2071-
8330.2022/15-3/5
P edic ion o business cycle o Poland
And ea Tkaco a
Facul y o Economics, Technical Uni e si y o Košice,
Nemco ej 32, 040 01 Kosice,
Slo ak Republic
and ea. ka[email p o ec ed]
ORCID 0000-0003-0268-009X
Bea a Ga u o a
Facul y o Managemen and Economics,
Tomas Ba a Uni e si y in Zlín,
Mos ní 5139, 760 00 Zlín,
Czech Republic
ga u o a@u b.cz
ORCID 0000-0002-0606-879X
Ka a ina Kelemen
Facul y o Economics,
Technical Uni e si y o Košice,
Nemco ej 32, 040 01 Kosice,
Slo ak Republic
ka a ina.kelemen@ uke.sk
ORCID 0000-0003-3540-1093
Abs ac . The pape is ocused on he cons uc ion o a new composi e indica o
in ended o p edic he economic cycle o Poland and i s compa ison wi h he
exis ing CLI used by in e na ional ins i u ions such as OECD and Eu os a . In
pa , his esea ch is also dedica ed o moni o ing he pa ial ad ance cyclical
indica o s ha make up he CLI componen s and hei changes o e ime. The
pape explo es 62 quali a i e and quan i a i e economic indica o s o Poland and
hei ela ionship o he de elopmen o mon hly GDP a cons an p ices in h ee
di e en ime pe iods: 2005 o 2021, 2010 o 2021, and 2016 o 2021. A modi ied
OECD me hod is used o selec he cyclical componen o ime se ies using he
Hod ick-P esco il e and subsequen ly employ c oss-co ela ion o he
a iables wi h he cyclical componen o GDP. The cons uc ed CLI can p edic
he e olu ion o he CLI one mon h ahead wi h a c oss-co ela ion le el o 0.879
unde equal weigh s and 0.877 unde di e en weigh s. Resea ch has shown ha
Recei ed:
Janua y, 2022
1s Re ision:
June, 2022
Accep ed:
Sep embe , 2022
DOI:
10.14254/2071-
8330.2022/15-3/5
Jou nal
o In e na ional
S udies
Scien i ic Pape s
© Founda ion
o In e na ional
S udies, 2022
© CSR, 2022
Jou nal o In e na ional S udies
Vol.15, No.3, 2022
66
he e is no signi ican change in he composi ion o he CLI o he p edic ion o
he economic cycle o Poland when using he es ablished me hodology.
Keywo ds: Business Cycle, Composi e Leading Indica o (CLI), GDP, C oss-
co ela ion, P edic ion.
JEL Classi ica ion: E32, E37.
1. INTRODUCTION
Moni o ing o he business cycle has been a he o e on o economic esea ch since he 1920s, mainly
due o he ins abili y o he economy and he subsequen G ea Dep ession in he 1930s. In he 1950s and
he 1960s, he business cycle seemed o be “dead”, bu he 1970s and he oil c ises b ough i back in o he
spo ligh and new economic heo ies came along ying o explain he causes o i s o igin (Be ge, 2011). A
new wa e o in e es in obse a ion and p edic ion o he economic cycle occu ed a he ime in he 21s
cen u y ha will go down in his o y as a pe iod ull o signi ican economic shocks ha o en could no be
eliably p edic ed. The deep nega i e e ec s o he inancial c isis in 2008, he subsequen wa e o he
ex eme public deb s o he Eu opean coun ies, he unexpec ed global onse o he co ona i us pandemic
in 2019, and he wa in Uk aine ha e aised he ques ion o whe he and o wha ex en we a e able o
p edic he possible u u e de elopmen o na ional economies.
Cu en ly, he e a e a ious app oaches o p edic ing business cycles using econome ic models (Cla k
& Ra azzolo, 2015; Fe a a, 2015; Ba houmi, 2016; Bjø nland, 2017; Pawę a, 2018; Kliuchnika a, 2022).
Some au ho s use speci ic gene al indexes (Gaweł & Głodowska, 2021; Wachi a, 2022; Ba os e al. 2021,
2022), howe e , globally he a en ion is ocused on p edic ions wi h he help o composi e leading
indica o s (CLIs). A Composi e Leading Indica o (CLI) is an index ha agg ega es he ime se ies
summa ising in o ma ion con ained in a numbe o key sho - e m economic indica o s known o be
associa ed wi h he business cycle (Zalewski, 2008). The economic cycle is ep esen ed by he de elopmen
o GDP, indus ial p oduc ion index o Composi e Coinciden Indica o (CCI) (A noš o á e al., 2011;
Baca ić e al., 2016). CLI p o ides quali a i e da a on sho - e m economic mo emen s. CLI d aws much
a en ion because i should be able o p edic he u u e s a es o economic ac i i y – when he economy is
going o swi ch om he expansion phase in o he con ac ion phase o ice e sa (V aná, 2008).
De e mina ion o u ning poin s is one o he main bene i s o using CLI (Baka ić e al, 2016). I is impo an
o no e ha CLIs o e in o ma ion abou he expec ed de elopmen o he economy, and hus open a
deba e on he implemen a ion o decisions in he public o p i a e sec o (Monni e al. 2017; Fabuš, 2017).
This is a sho - e m es ima e o he u u e economic si ua ion, bu i can poin o impo an economic
changes ha a e he u ning poin s o he economic cycle (K amoliš, 2015; Mazu , 2017, Döpke e al. 2017).
In addi ion o he p edic ion i sel , CLI can also be used o analyse he pe o mance o economies and hus,
o compa e hei pe o mance wi hin di e en g oups (Sal elli, 2006).
The composi e leading indica o s also possess own disad an ages and limi s ha a e isible in hei
p ac ical use ha is common o he composi e indica o s (see Djogo & S anišić, 2016; Roszko-Wój owicz
& Białek, 2016; Šego a e al., 2017). Among he undamen al issues, i is possible o conside he inco ec
selec ion o he e e ence se ies o an insu icien ly la ge g oup o sub-indica o s. When using CLI, i is
mainly abou he a ailabili y o he alse signals and misin e p e a ion o he esul s (Saisana & Ta an ola,
2002; Eu os a , 2017).
Fo he economic policy measu es de i ed om he CLI p edic ion, i is necessa y o be cau ious and
o ake in o accoun he impac o se e al ac o s, such as he de elopmen o CLI pa ial indica o s, he
And ea Tkaco a, Bea a Ga u o a,
Ka a ina Kelemen
P edic ion o business cycle o Poland
67
o e all na u e o he economy, he causes o nega i e de elopmen s (p oblems o he banking sec o , public
inances, de elopmen s in he c edi ma ke , he quali y o business en i onmen , economic de elopmen s
in he o he coun ies depending on he openness o he economy, and he o he aspec s. (T a kina, 2015;
Jan o-D ozdowska & Majewska, 2016; Fuinhas e al., 2016; Sachpazidu-Wójcicka, 2017; Dobeš e al. 2017;
Ko sakienė e al. 2017). One o he mos impo an ac o s ha can shi signi ican ly p edic ion esul s is
he high sha e o in o mal economy (Digdowiseiso & Sugiyan o, 2021; Jo o ic, 2021; Mishchuk e al., 2018)
as well as unce ain y in inno a ions and knowledge-based ac i i ies de elopmen (Ebong & Babu, 2020;
Oliinyk e al., 2021; Wasiluk & Gine ičius, 2020). I is equally impo an ha some iscal policy decisions
can only be posi i e o he economy in he sho e m, bu in he long un, hey may be coun e p oduc i e
(Ko sakienė e al. 2015; Ciegis e al., 2015; Laj kepo á, 2016; Vochozka e al. 2020, 2021).
Due o he ad an ages o CLI, he na ional s a is ical o ices, he na ional banks, as well as he speci ic
en e p ises, p ima ily wi h an indus y o ien a ion, a e de o ed o explo e hei cons uc ion and
subsequen ly, o analyse hese poin s. CLI possesses an impo an posi ion in he sho - e m economic
o ecas s o he Eu opean coun ies, he Uni ed S a es o Ame ica, and Japan. A he in e na ional le el, he
o ganisa ions such as he OECD and Eu os a pay a signi ican a en ion o hem. They p e e a di e en
CLI design app oach and hence, hey use mainly he same CLI composi ion o he selec ed coun y o a
long ime. This esul s in quali y p edic ion capabili ies o CLI no being achie ed o some economies. Fo
his eason, i is necessa y o e i y he p edic i e capabili ies o exis ing CLIs and, i necessa y, c ea e new
composi e p e-hea indica o s. Fo economies, which ha e unde gone he signi ican changes in he pas , i
can be assumed ha he composi ion o hei CLI will change o e ime. The ansi ional economies can be
conside ed such coun ies. Fo he pu poses o his pape , Poland was selec ed as he coun y unde s udy
due o he insu icien p edic i e abili y o CLI by he OECD and Eu os a and his is analysed in mo e
de ail in he gi en pape . The aim o he p esen con ibu ion is o c ea e an own composi e leading indica o
o he business cycle o Poland and compa e i wi h he exis ing CLIs. The esea ch ques ion is also aised
whe he he e eally is a change in he composi ion o he CLI in he case o Poland o e ime.
2. LITERATURE REVIEW
The economis s a he na ional and in e na ional le el a e engaged in he cons uc ion o he CLI o
he economic cycle o Poland. The bes -known in e na ional me hodologies o he cons uc ion o Poland's
CLI a e ep esen ed by he OECD and Eu os a p ocedu es. Thei modi ica ions can be seen in he s udies
o he au ho s such as Bandholz (2005), Zalewski (2009), Jakubíko á e al. (2014), and V aná (2018).
The me hodology o he OECD and Eu os a is based on he g ow h cycle, while he ime se ies can
be di ided in o he andom, end, seasonal, and cyclical componen s (T imbu , 2006). Fo u he
in es iga ion, he cyclical componen is selec ed om he ime se ies. The i s impo an s ep is he selec ion
o a e e ence se ies ha ep esen s he economic cycle o he gi en economy. The s udies o e se e al
op ions o a selec ion o he e e ence se ies o Poland. The mon hly index o he indus ial p oduc ion
(o manu ac u ing p oduc ion) is he mos commonly used measu e o economic ac i i y (Bandholz, 2005).
The i s eason is ha i is a ailable p omp ly and on a mon hly basis in con as o GDP. Secondly, i
cons i u es he mos cyclical subse o he whole economy. Mo eo e , o many coun ies i was ound ha
cyclical p o iles o GDP and IIP a e s ongly ela ed. An ob ious disad an age o employing GDP ins ead
o IIP is ha GDP is e y o en e ised by he Cen al S a is ical O ice and i is a subjec o signi ican
changes (OECD, 2006). Acco ding o he OECD me hodology, he IIP was applied o cons uc CLI un il
2012 o Poland and om 2012 i was a mon hly ime se ies o GDP a cons an p ices (Fulop & Gyomai,
2012). GDP is also p e e ed by Eu os a (2017), Bandholz (2005), Zalewski (2008), and Jakubíko á e al.
(2014). On he o he hand, au ho s such as A is e al. (2004) and V aná (2018) p e e he use o IIP. In
Jou nal o In e na ional S udies
Vol.15, No.3, 2022
68
hei opinion, GDP o Poland shows oo li le cyclical a ia ion and is hus no he app op ia e measu e
o moni o ing business cycle luc ua ions.
A b ie compa ison o he elemen a y di e ences in he cons uc ion o he OECD CLI and Eu os a
composi e indica o s is p esen ed in Table 1.
Table 1
Compa ison o in e na ional me hodologies o CLI cons uc ion
C i e ia
Eu os a
OECD
Type o business cycle
G ow cycle
G ow cycle
Coun ies
Chosen coun ies o Eu opean
Union, Eu ozone
Membe s o OECD, a ew
nonmembe coun ies
T end de e mina ion
Hod ick P esco Fil e
Ch is iano Fi zge ald Fil e
Re e ence se ies
GDP
Index o indus ial p oduc ion ( o
2012)
GDP (since 2012)
Rela ionship be ween e e ence
se ies and indica o s
C oss co ela ion
C oss co ela ion
Type o da a
Quali a i e da a
Quan i a i e and quali a i e da a
Pe iod o da a
Mon hly da a
Mon hly da a
Wages o componen s
Digge en wages o all componen s
Same wages o all componen s
Sou ce: Au ho s’ esul s.
The p edic ion possibili ies o he cyclical de elopmen o Poland's economy a e discussed in he
s udies by D ozdowicz-Bieć (2001), Ma kowski (2002), Nilsson (2006), Ga cza czyk & Skikiewicz (2011),
Lena e al. (2016) o Mazu (2017). Ano he au ho Bandholz (2005) applies GDP as he e e ence se ies
and hus, he c ea es he CLI o Poland and Hunga y om qua e ly da a o he pe iod 1994 o 2004.
Employing he linea and non-linea dynamic ac o modelling app oaches, he inds o he bo h coun ies
ha a pa simonious speci ica ion ha combines he na ional business cycle indica o s, he se ies e lec ing
ade olumes, and he sup ana ional business expec a ions making o he mos eliable business cycle
leade s. The composi e leading indica o s signi ican ly cause a GDP g ow h, while he es ima ed Ma ko -
swi ching p obabili ies o being in a ecessiona y s a e ag ee well wi h a p io i de e mined cycle ch onologies.
The esul ing CLI o Poland is composed o a sys em o he equal and di e en weigh s and i con ains he
six componen s.
Zalewski (2009) applies he modi ied o m o he OECD me hodology and hus, he chooses mon hly
IIP da a o he pe iod 1992 o 2007 as he e e ence se ies. Based on he analysis o he 15 economic
indica o s o Poland, he compiles he CLI, which consis s o six esul ing componen s. He compa es he
cons uc ed CLI wi h he de elopmen o ESI and Ma kowski (2002), e alua ing ha his cons uc ed CLI
indica es ewe alse signals and he e o e, i p o ides a be e sho - e m p edic ion.
Jakubíko a e al. (2014) explo ed he de elopmen o he economic cycle o he V4 coun ies wi h a
e e ence se ies o GDP a cons an p ices. Using he analysis o he qua e ly da a o he pe iod 2005 o
2011, hey poin ed ou he sho comings in he OECD p edic i e abili y o Poland. In he explo ed pe iod,
he highes alue o he OECD CLI c oss-co ela ion was 0.489, while Eu os a ESI s ood a a le el o
0.798. Subsequen ly, acco ding o he modi ied OECD me hodology, a CLI was compiled o Poland oo,
and i consis ed o he six componen s also. Fo he V4 coun ies, he CLI has been shown o ha e a
di e en composi ion o each coun y.
V aná (2018) applied an inno a i e app oach o he cons uc ion o he CLI, which consis s in
including a iables om o he coun ies among he selec ed cyclical indica o s. To compile CLI o Poland,
i ollows he de elopmen o indica o s in Aus ia, Ge many, he Czech Republic and Slo akia o he
pe iod 1996 o 2016 wi h he e e ence se ies o GDP. He a gues ha he economies o some coun ies a e
And ea Tkaco a, Bea a Ga u o a,
Ka a ina Kelemen
P edic ion o business cycle o Poland
69
small and hei economic si ua ion is o en ela ed o he de elopmen o he business cycle in o he
coun ies. Acco ding o he s udy, he esul ing CLI o Poland ac ually achie es he bes p edic i e
capabili ies when inco po a ing indica o s om he moni o ed coun ies.
The a ailable CLIs o Poland we e compa ed based on selec ed c i e ia such as a CLI composi ion,
lead size and CLI c oss-co ela ion alue (Table 2).
Table 2
Compa ison o he cons uc ion o he composi e indica o s o Poland
Au ho
Obse ed
pe iod
Indica o cons uc ion
Time
pe iod
ad ance
C oss-
co ela ion
alue
Bandholz
(2005)
1994-2004
(qua e ly
da a)
CLI ( e e ence se ies GDP):
Desc ibes demand endency in manu ac u ing,
Nominal e ec i e exchange a e, Business
expec a ions o Wes e n Eu ope
1 Q
0.75
Zalewski
(2009)
1992-2007
(mon hly
da a)
CLI ( e e ence se ies IPP):
Na ow Money (M1) Index 2000=100 SA,
Manu ac u ing indus y (Selling p ices, Fu u e
endency), Manu ac u ing indus y (P oduc ion
Tendency), Sho - e m in e es a es, Manu ac u ing
indus y (P oduc ion, Fu u e Tendency), Ne ade in
goods ( alue) in billions o US dolla s SA
-
0.788
Jakubíko á e
al. (2014)
2005-2011
(qua e ly
da a)
CLI ( e e ence se ies GDP):
P oduc ion in manu ac u ing indus y, 2005 = 100
Mone a y agg ega e M1, 2005 = 100 Wa szawski
Indeks Giełdowy, 2005 = 100, Indus y u no e ,
(in e media e p oduc and capi al goods), domes ic
ma ke , 2005 = 100 Indica o o con idence in he
cons uc ion indus y
2 Q
0.886
V aná (2018)
1996-2016
(mon hly
da a)
CLI ( e e ence se ies GDP):
9 in e na ional componen s (Aus ia, Czech
Republic, Ge many, Slo akia) and 6 na ional
componen s:
Business endency su eys (cons uc ion), Sha e
P ices, Business endency su eys (manu ac u ing),
In e na ional T ade (Impo s, goods), Business
endency su eys (se ices- business si ua ion),
Business endency su eys (se ices - con idence
indica o s)
3 M
0.790
OECD* (2022)
2005-2021
2010-2021
2016-2021
(mon hly
da a)
CLI ( e e ence se ies GDP):
Real e ec i e exchange a es - CPI Based
(2015=100) in e ed 3-mon h WIBOR (% pe
annum) in e ed Manu ac u ing - P oduc ion:
endency (% balance) Job acancies: un illed
(numbe ) P oduc ion o coal ( onnes)
10 M
9 M
9 M
0.442
0.578
0.588
Eu os a *
(2022)
2005-2021
2010-2021
2016-2021
(mon hly
da a)
CLI ( e e ence se ies GDP):
Con idence indica o in indus y, Con idence
indica o in he se ice, Consume con idence
indica o , Con idence indica o in cons uc ion,
Con idence indica o in e ail
1 M
0 M
0 M
0.756
0.811
0.921
No e: * own calcula ions
Sou ce: Au ho s’ esul s

Jou nal o In e na ional S udies
Vol.15, No.3, 2022
70
In he case o CLI OECD and ESI Eu os a , he alues we e calcula ed by he au ho s o he pape .
By compa ing he obse ed s udies, he di e ence in he p edic i e abili ies o CLI is no iceable. The eason
is mainly di e ences in he CLI calcula ion me hodology, bu also in he ype and pe iodici y o moni o ed
a iables and he leng h o he ime se ies. The p edic i e capabili ies o he OECD and Eu os a CLIs,
which we e calcula ed by he au ho s o his s udy om h ee di e en leng h ime se ies wi h unchanged
CLI composi ion, a e in e es ing. In he case o he OECD, a p edic ion pe iod o up o 9 o 10 mon hs
was ound, bu wi h a low c oss-co ela ion alue o 0.442 o 0.588. This means ha wi h his composi ion,
he CLI gi es many alse signals abou he changes in he Polish business cycle. On he con a y, he ESI
showed a high co ela ion alue om 0.756 o 0.921, bu a he ime o he coincidence o a mon h in
ad ance.
3. METHODOLOGY
The main goal o he pape is he cons uc ion o a new composi e leading indica o (CLI) designed o
p edic he de elopmen o he business cycle in Poland. A pa ial objec i e is o de e mine whe he he
leng h o he ime se ies signi ican ly a ec s he g oups o he cyclical leading indica o s ha can o m CLIs.
Fo c ea ion o CLI, a modi ied OECD me hodology is employed. I is based on a cons uc ion wi h he
g ow h cycle (De V oey & Pensie oso, 2006). This is mo e app op ia e o apply in he case o he ansi ional
economies such as Poland (Czesaný e al., 2007). The mon hly GDP indica o a cons an p ices is selec ed
as he e e ence se ies ha is gene ally conside ed he b oades indica o o economic ac i i y (Czesány &
Jeřábko á, 2009a). Only i s cyclical componen is used in he analyses (Czesány & Jeřábko á, 2009b; As ol i
e al., 2016).
The indica o s, whose ela ion o GDP is moni o ed include, he 62 quan i a i e and quali a i e
indica o s om he ields o indus y, se ices, e ail, cons uc ion, o eign ade, labo ma ke , mone a y
agg ega es, s ock indices, con idence indica o s, consume expec a ions as well as he GDP componen s
hemsel es. The da a sou ces a e he da abases o he OECD, Eu os a and he Polish S a is ical O ice. All
he a ailable ime se ies om Janua y 2005 o No embe 2021 wi h mon hly pe iodici y a e included in he
analysis. The ela ionship o he a iables o GDP in he pe iods 2005 o 2021, 2010 o 2021, and 2016 o
2021 is moni o ed. The eason is he esul s o he published esea ch o Slo akia, which con i med he e
is a change in he composi ion o he CLI o e ime o ha he leng h o he obse ed pe iod a ec s he
composi ion o he CLI. The ecommenda ion is o moni o changes in he composi ion o he CLI a leas
a he i e-yea in e als, o a e majo economic changes in he coun y and he eby, o minimise he
numbe o he alse signals (Tkáčo á & Kišo á, 2018).
The ime se ies is ini ially seasonally adjus ed by he me hod o balancing employing he seasonal
indices and hen, he end is emo ed applying he Hod ick-P esco il e (HP il e ). The HP il e is a
commonly employed ool o de ending. I is a mos a ou able ex ac o o a end ha is s ochas ic bu
mo es smoo hly o e ime and is unco ela ed wi h he cycle (Ko acic & Vilo ic, 2017). Fo =1,2,3… he
end componen Y* is compu ed, and  is chosen o minimise:
∑(𝑌
𝑡
𝑇
𝑡=1 − 𝑌
𝑡∗)2+ λ ∑[(𝑌
𝑡+1
∗− 𝑌
𝑡∗
𝑇−1
𝑡=2 ) − (𝑌
𝑡∗− 𝑌
𝑡−1
∗)]2 (1)
To ge op imal esul s o de ending, i has been sugges ed o choose λ=1600 o qua e ly da a and λ
=14 400 o mon hly da a (Schilch , 2005). An ad an age o he HP me hod is ha no es ic ion on he
leng h o ime se ies is imposed. Ne e heless, he e is a equi emen ha be o e p oceed wi h he HP il e
one should seasonally adjus ed each se ies. The end i sel is no e y in e es ing in he analysis o cyclical
And ea Tkaco a, Bea a Ga u o a,
Ka a ina Kelemen
P edic ion o business cycle o Poland
71
beha iou . The e o e, he es o s udy was done wi h cyclical componen s o each se ies (Nilsson & B une ,
2006).
The c oss-co ela ion in a way o he Pea son's co ela ion coe icien wi h he wo o wa d shi s and
he i e backwa d shi s is applied in o de o de e mine he ela ionship be ween he a iables. Acco ding
o he alue o he co ela ion coe icien and he size o he ad ance, i is possible o c ea e g oups o
cyclical and an icyclical indica o s acco ding o he pa e n o Table 3.
Table 3
C i e ia o iden i ica ion o he cyclical beha iou o he indica o s o Poland
Indica o ype
The highes absolu e c oss-
co ela ion alue
Time o each he highes absolu e
c oss-co ela ion alue
Cyclic
Ea ly
> 0,55
( +1, +2)
O e due
> 0,55
( -5, -1)
Concu en
> 0,55
An icyclical
≤ 0,55
no signi ican
Sou ce: Kľúčik, Haluška (2008).
A g oup o he leading indica o s is a subjec o a mo e de ailed analysis o he compila ion o Poland's
CLI. Subsequen ly, he selec ion and sco ing me hod is applied, which economic and s a is ical signi icance
and s a is ical quali y a e e alua ed acco ding o. The maximum numbe o he poin s ha he a iables can
each is gi en in Table 4.
Table 4
Sco ing c i e ia o he selec ion o he leading cyclical indica o s
Economic signi icance
(maximum 10 poin s)
S a is ical signi icance
(maximum 30 poin s)
S a is ical quali y
(maximum 10 poin s)
Economic
in e p e a ion in
ela ion o he
business cycle
10 poin s
Pea son's co ela ion
coe icien
15 poin s
Time a ailabili y
5 poin s
Numbe o ad ance
mon hs
15 poin s
Upda e
5 poin s
Sou ce: Au ho s’ esul s
Due o he di e en uni s o he pa ial indica o s, hei no malised alues ob ained by using he
s anda diza ion me hod (OECD, 2008) a e used in he composi ion o he CLI. A sys em o he equal and
di e en weigh s is employed. The ela ionship o calcula ing equal weigh s looks like as ollows:
𝜔𝑖=1
𝑛 (2)
whe e i s ands o he weigh o i h componen a n is he numbe o he pa ial indica o s en e ing he
CLI. The absolu e alues o he co ela ion coe icien s a e applied in o de o de e mine he di e en
weigh s.
The ela ionship o calcula ion o he di e en weigh s is exp essed subsequen ly:
𝜔𝑖= 𝑟𝑖
∑𝑟𝑖
𝑛
𝑖=1 (3)
whe e i s ands o he weigh o i h componen , is he absolu e alue o he co ela ion coe icien o
i h componen a he ime o ad ance and n is he numbe o he pa ial indica o s en e ing he CLI.
The sum o he no malised alues mul iplied by he weigh s c ea es he basemen o cons uc ion o
he CLI assembly equa ion.
𝐶𝐿𝐼𝑡= ∑𝜔𝑖
𝑛
𝑖=1 ∗ 𝑦𝑖,𝑡 (4)
𝜔 – a a iable weigh alue,
y – a alue o he no malised cyclic componen o he a iable a ime,
Jou nal o In e na ional S udies
Vol.15, No.3, 2022
72
n – a numbe o he pa ial indica o s en e ing he CLI.
4. EMPIRICAL RESULTS AND DISCUSSION
The basic s ep o he c ea ion o CLI o Poland is cha ac e ised by he selec ion o a e e ence se ies
ha ep esen s he economic cycle. Th ough he analysis o he ela ionship be ween he cyclical componen
o GDP and PPI, i was demons a ed ha PPI shows a mon h ad ance wi h a c oss-co ela ion alue o
0.789 o he pe iod 2005 o 2021 and 0.827 o he pe iod 2016 o 2021. This means ha he PPI ep esen s
he leading indica o o GDP and hus, i is able o o m one o he componen s o he CLI. I is no
app op ia e o selec i as a e e ence line. In he case o Poland, i is mo e sui able o use he cyclical
componen o GDP.
Selec ion o he cyclical indica o s o Poland
Acco ding o he desc ibed me hodology, he cyclical ela ionship o all he 62 selec ed a iables o
GDP was analysed. Th ough calcula ion o he c oss-co ela ions o he cyclical componen s associa ed
wi h he a iables o he cyclical componen o GDP, he 17 indica o s we e selec ed ha demons a ed he
p ope ies o he leading cyclical indica o s acco ding o he de ini ion gi en in he me hodological pa o
he pape . The highes c oss-co ela ion alue was achie ed o he le side o and a he same ime, he
second highes c oss-co ela ion alue was g ea e han 0.55. Table 5 demons a es he esul s.
Table 5
Ad ance indica o s
Indica o
2005–2021
2010–2021
2016–2021
ad ance
co ela ion
ad ance
co ela ion
ad ance
co ela ion
Indus ial p oduc ion index
1 M
0.798
1 M
0.802
1 M
0.827
S ock p ice index
2 M
0.524
2 M
0.533
1 M
0.664
To al indus y p oduc ion index
1 M
0.807
1 M
0.813
1 M
0.838
To al p oduc ion in he p ocessing
indus y index
1 M
0.775
1 M
0.774
1 M
0.796
To al p oduc ion in he p ocessing
indus y index – in e media e
p oduc
1 M
0.732
1 M
0.756
1 M
0.761
To al in es men goods p oduc ion
index
1 M
0.745
1 M
0.751
1 M
0.757
To al e ail olume index
1 M
0.595
1 M
0.622
1 M
0.745
P oduc ion in he p ocessing
indus y
1 M
0.667
1 M
0.723
1 M
0.813
Consume indus y employmen
1 M
0.599
1 M
0.665
1 M
0.747
P ocessing indus y con idence
indica o
1 M
0.714
0 M
0.786*
0 M
0.857*
Cons uc ion indus y employmen
1 M
0.619
1 M
0.719
1 M
0.796
Re ail business si ua ion
1 M
0.706
1 M
0.750
1 M
0.801
Re ail employmen
1 M
0.702
1 M
0.805
1 M
0.834
O de in en ion o demand in e ail
1 M
0.708
1 M
0.760
1 M
0.811
Se ices – demand de elopmen
1 M
0.760
1 M
0.788
1 M
0.862
Se ices – employmen
1 M
0.686
1 M
0.766
1 M
0.813
Dow Jones eu o s oxx 50 P ice
Index
1 M
0.483*
1 M
0.479*
1 M
0.667
* he indica o is assigned o he ad ance indica o s acco ding o he ad ance pe iod leng h and no he c oss-
co ela ion alue
Sou ce: Au ho s’ esul s.
And ea Tkaco a, Bea a Ga u o a,
Ka a ina Kelemen
P edic ion o business cycle o Poland
73
Th ee ime pe iods we e moni o ed in o de o e i y he assump ion ha he e is a change in he
leading cyclical indica o s h oughou he ime ha ep esen po en ial componen s o he CLI. In he case
o Poland, only a small change in he na u e o hese indica o s was con i med. Ou o he 62 a iables, 17
indica o s we e included among he leading cyclical indica o s. Subsequen ly, only he 2 ou o he 17
indica o s changed hei na u e as a leading cyclical indica o . The i s one was an indica o o con idence
in he manu ac u ing indus y, which o e ime changed om a leading indica o o a concu en indica o ,
and he second was he Dow Jones Eu o S oxx 50 p ice index ha acqui ed he cha ac e is ics o a leading
indica o o e ime. Fo he 15 indica o s, i has been p o en ha hey show a be e quali y lead wi h a
sho e ime pe iod, and hus he alue o hei co ela ion coe icien inc eases. O e all, i can be assumed
ha in he case o Poland, he e is no signi ican change o e a ime pe iod in he g oup o he leading
cyclical indica o s. Acco ding o he es ablished me hodology, i is possible o compile a high-quali y CLI
he e o e, whose composi ion would mee he p edic ion o he economic cycle o Poland e en o se e al
yea s. Howe e , wi h sho e ime pe iods, he alue o he c oss-co ela ion g ew, which means ha he
selec ed indica o s a e sui able o c ea ing a CLI o he cu en p edic ion. Fo his eason, he subsequen
in es iga ion is de o ed o he ime se ies o he 2016 o 2021 pe iod.
Fo a de ailed selec ion o he a iables ha c ea e he CLI, a selec ion and a sco ing me hod is applied
o all he 17 lead indica o s (Table 6), while he maximum alue, which he indica o can each, is 50 poin s
(Kľúčik, 2009b). The a iables om he same economic a ea, which achie ed he highes numbe o poin s
a e an applica ion o a selec ion and a sco ing me hod, we e eligible o u he analysis. This educed he
selec ion down o he 10 leading cyclical indica o s.
Table 6
Resul s o he selec ion and sco ing me hod o he ad ance cyclical indica o s
Indica o
Maximum
poin s
Indica o
Maximum poin s
To al indus y p oduc ion index
45
To al e ail olume index
40
Indus ial p oduc ion index
44
Re ail employmen
39
To al in es men goods
p oduc ion index
43
O de in en ion o demand in e ail
39
S ock p ice index
42
P ocessing indus y con idence
indica o employmen
39
To al p oduc ion in he p ocessing
indus y index – in e media e
p oduc
42
P ocessing indus y con idence
indica o
38
To al p oduc ion in he p ocessing
indus y index
41
Se ices – demand de elopmen
38
To al p oduc ion in he p ocessing
indus y
41
Dow Jones eu o s oxx 50 P ice Index
38
Re ail business si ua ion
41
P ocessing indus y con idence
indica o
25
Se ices – employmen
41
Sou ce: Au ho s’ esul s
Cons uc ion o a composi e lead indica o o Poland
Du ing he cons uc ion o he CLI, he e we e he 7 op ions o he composi ion o he CLI wi h each
addi ional possibili y ha ing one less indica o wi h he lowes numbe o he poin s. CLI 10 (a he same
weigh s) o CLI A (a di e en weigh s) we e composed o he 10 indica o s and CLI 4 o he 4 indica o s
ha ep esen ed he minimum numbe . CLI nume ical designa ion is applied o he same indica o weigh s
and a le e designa ion is used employed o he di e en CLI componen weigh s. Table 7 demons a es
he esul s o he c oss-co ela ions wi h he ad ance olume.
Jou nal o In e na ional S udies
Vol.15, No.3, 2022
80
Šego a, A., Tomljano ić, M., & Huđek, I. (2017). Con empo a y app oaches o measu ing compe i i eness – he case
o EU membe s a es. Jou nal o Economics and Business, 35 (1), 123-150.
h ps://doi.o g/10.18045/zbe i.2017.1.123.
Tkáčo á, A., & Kišo á, V. (2017). Zmeny zloženia kompozi ného p eds iho ého indiká o a Slo enska čase. Poli ická
Ekonomie, 65(5), 583-600. h ps://doi.o g/10.18267/j.polek.1163.
T a kina, I. (2015). Expo and GDP G ow h in Li huania: Sho - un o Middle- un Causali y? En ep eneu ship and
Sus ainabili y, Issues 3(1), 74-84. h ps://doi.o g/10.9770/ jesi.2015.2.4(7).
T imbu , T. M. (2006). De ending economic ime se ies: a Bayesian gene aliza ion o he Hod ick-P esco il e .
Jou nal o Fo ecas ing, 25(4), 247-273. doi:10.14254/2071- 789x.2015/8-2/10.
Vochozka, M., Kalino á, E., Gao, P. & Smolíko á, L. (2021). De elopmen o coppe p ice om July 1959 and
p edic ed de elopmen ill he end o yea 2022. Ac a Mon anis ica Slo aca, 26(2), 262-280.
h ps://doi.o g/10.46544/AMS. 26i2.07
Vochozka, M., Ho ák, J., K ulický, T. & Pa dal, P. (2020). P edic ing u u e B en oil p ice on global ma ke s. Ac a
Mon anis ica Slo aca, 25(3), 375-392. h ps://doi.o g/10.46544/AMS. 25i3.10
V aná, L. (2018). On ex ending composi e leading indica o s by in e na ional economic se ies. S a is ika. 98(2). 113-13
Wachi a, E.W. (2022). Analysis o Aus ia's En ep eneu ial Ecosys em Based on he Gei App oach. In e na ional
Jou nal o En ep eneu ial Knowledge, 10(1), 123-136. h ps://doi.o g/10.37335/ijek. 10i1.156
Wasiluk, A., & Gine ičius, R. (2020). P o-inno a i e mo i es o es ablishing coope a ion by en e p ises: An empi ical
s udy in Poland. Economics and Sociology, 13(2), 258-278. h ps://doi.o g/10.14254/2071-789X.2020/13-2/17
Zalewski, K. (2009). Fo ecas ing Tu ning Poin s wi h Composi e Leading Indica o s - he Case o Poland. Ekonomia
Jou nal, 24, 61-9.

And ea Tkaco a, Bea a Ga u o a,
Ka a ina Kelemen
P edic ion o business cycle o Poland
81
ANNEX
Figu e 1. De elopmen o he cyclical componen o GDP and CLI wi h he equal and di e en
weigh s
Sou ce: Au ho s’ esul s
Figu e 2. Compa ison o he de elopmen o he cyclical componen o GDP, CLI 5 ( he equal
weigh s), CLI OECD, ESI
Sou ce: Au ho s’ esul s
-8,00
-6,00
-4,00
-2,00
0,00
2,00
4,00
Jan-2016
Ma -2016
May-2016
Jul-2016
Sep-2016
No -2016
Jan-2017
Ma -2017
May-2017
Jul-2017
Sep-2017
No -2017
Jan-2018
Ma -2018
May-2018
Jul-2018
Sep-2018
No -2018
Jan-2019
Ma -2019
May-2019
Jul-2019
Sep-2019
No -2019
Jan-2020
Ma -2020
May-2020
Jul-2020
Sep-2020
No -2020
Jan-2021
Ma -2021
May-2021
Jul-2021
Sep-2021
No -2021
Cyclical componen o GDP CLI 5 CLI F
-5,00
-4,00
-3,00
-2,00
-1,00
0,00
1,00
2,00
-8,00
-6,00
-4,00
-2,00
0,00
2,00
4,00
Jan-2016
Ma -2016
May-2016
Jul-2016
Sep-2016
No -2016
Jan-2017
Ma -2017
May-2017
Jul-2017
Sep-2017
No -2017
Jan-2018
Ma -2018
May-2018
Jul-2018
Sep-2018
No -2018
Jan-2019
Ma -2019
May-2019
Jul-2019
Sep-2019
No -2019
Jan-2020
Ma -2020
May-2020
Jul-2020
Sep-2020
No -2020
Jan-2021
Ma -2021
May-2021
Jul-2021
Sep-2021
No -2021
cyclical componen o GDP CLI OECD ESI CLI 5