scispace - formally typeset
Search or ask a question
Journal ArticleDOI

A Measure of Comovement for Economic Variables: Theory and Empirics

TL;DR: In this article, a measure of dynamic comovement between (possibly many) time series and names it cohesion is defined in the frequency domain and is appropriate for processes that are costationary, possibly after suitable transformations.
Abstract: This paper proposes a measure of dynamic comovement between (possibly many) time series and names it cohesion. The measure is defined in the frequency domain and is appropriate for processes that are costationary, possibly after suitable transformations. In the bivariate case, the measure reduces to dynamic correlation and is related, but not equal, to the well known quantities of coherence and coherency. Dynamic correlation on a frequency band equals (static) correlation of bandpass-filtered series. Moreover, long-run correlation and cohesion relate in a simple way to co-integration. Cohesion is useful to study problems of business-cycle synchronization, to investigate short-run and long-run dynamic properties of multiple time series, and to identify dynamic clusters. We use state income data for the United States and GDP data for European nations to provide an empirical illustration that is focused on the geographical aspects of business-cycle fluctuations.

Summary (1 min read)

Jump to:  and [Summary]

Summary

  • Wkh phdvxuh lv gh qhg lq wkh iuhtxhqf| grpdlq dqg lv dssursuldwh iru surfhvvhv zklfk duh frvwdwlrqdu|/ srvvleo| diwhu vxlwdeoh wudqv0 irupdwlrqv1.
  • Frkhvlrq lv xvhixo wr vwxg| sureohpv ri exvlqhvv f|foh v|qfkurql}dwlrq/ wr lqyhv0 wljdwh vkruw0uxq dqg orqj0uxq g|qdplf surshuwlhv ri pxowlsoh wlph vhulhv/ wr lghqwli| g|qdplf foxvwhuv1.
  • Zh xvh vwdwh lqfrph gdwd iru wkh XV dqg JGS gdwd iru Hxurshdq qdwlrqv wr surylgh dq hpslulfdo looxvwudwlrq irfxvhg rq wkh jhrjudsklfdo dvshfwv ri exvlqhvv f|foh xfwxdwlrqv1 MHO Fodvvl fdwlrq=.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

i@thi Lu L4Li4i?| uLh ,UL?L4U V@h@M*itG
AiLh) @?_ ,4ThUt
ht|LTi hL
@hL 6Lh?
|
wUhi3@ +iU*?
}
Devwudfw
Nh|zrugv=
Exvlqhvv f|foh/ Vhfwrudo frpryhphqwv/ frkhuhqfh/ jhrjudsk|1
Wklv sdshu sursrvhv d phdvxuh ri g|qdplf frpryhphqw ehwzhhq
+
srvvleo| pdq|
,
wlph vhulhv dqg qdphv lw
frkhvlrq
1 Wkh phdvxuh lv ghqhg lq wkh iuhtxhqf| grpdlq
dqg lv dssursuldwh iru surfhvvhv zklfk duh frvwdwlrqdu|/ srvvleo| diwhu vxlwdeoh wudqv0
irupdwlrqv1 Lq wkh elyduldwh fdvh/ wkh phdvxuh uhgxfhv wr
g|qdplf fruuhodwlrq
dqg lv
uhodwhg/ exw qrw htxdo/ wr wkh zhoo nqrzq txdqwlwlhv
frkhuhqfh
dqg
frkhuhqf|
1 G|qdplf
fruuhodwlrq rq d iuhtxhqf| edqg htxdov
+
vwdwlf
,
fruuhodwlrq ri edqg0sdvv owhuhg vhulhv1
Pruhryhu/ orqj uxq fruuhodwlrq dqg frkhvlrq uhodwh lq d vlpsoh zd| wr frlqwhjudwlrq1
Frkhvlrq lv xvhixo wr vwxg| sureohpv ri exvlqhvv f|foh v|qfkurql}dwlrq/ wr lqyhv0
wljdwh vkruw0uxq dqg orqj0uxq g|qdplf surshuwlhv ri pxowlsoh wlph vhulhv/ wr lghqwli|
g|qdplf foxvwhuv1 Zh xvh vwdwh lqfrph gdwd iru wkh XV dqg JGS gdwd iru Hxurshdq
qdwlrqv wr surylgh dq hpslulfdo looxvwudwlrq irfxvhg rq wkh jhrjudsklfdo dvshfwv ri
exvlqhvv f|foh xfwxdwlrqv1
MHO Fodvvlfdwlrq=
H6/ F4
LhhitTL?_i?Ui __hittG ht|LTi hL c ,+,5c Df @i 6# +LLtii*|c h i**it
fDf i4@*G UUhL 9*M@UMi
Xqlyhuvlwh Oleuh gh Eux{hoohv/ HFDUHV
|
Xqlyhuvlw| ri Prghqd dqg FHSU
}
Xqlyhuvlwh Oleuh gh Eux{hoohv/ HFDUHV dqg FHSU1

W?|hL_U|L?
Aihi @hi ui i4ThU@* hi*@|L?t U @i |i t|@|t Lu Rt|)*3i_ u@U|t ? iUL?L4Ut
?i Lu |i4 t |@| 4@UhLiUL?L4U @}}hi}@|it UL4Li At LMtih@|L? @t Mii? |i
tLhUi Lu tTiU*@|L? Lu iUL?L4U |iLh) t?Ui |t Mh| W? 4L_ih? |iLhit Lu |i Mt?itt
U)U*i TiLT*i @i @t!i_ i|ih UL4Li4i?|t U@? Mi i T*@?i_ M) *@h}i @}}hi}@|i tLU!tc
4L?i|@h) Lh hi@*c Lh i|ih @? i T*@?@|L? tL*_ Mi uL?_ ? ?L?*?i@h ThLT@}@|L?
4iU@?t4t ,ih) 4@UhLiUL?L4U |i |MLL! t|@h|t M) @ t|@|i4i?| L? UL4Li4i?|t Mi|ii?
@}}hi}@|it OLiihc T@h@_L U@**)c |t t L?i Lu |i u@U|t |@| t *i@t| i** _LU4i?|i_
@?_ L? @| |ihi t 4Lhi UL?utL? Lu 4i@??} @?_ |ih4?L*L}) RL4Li4i?| t @
*Lti |ih4c TLttM*) _itUhM?} _gihi?| Ti?L4i?@ @?_c UL?ti^i?|*)c | 4@?) _gihi?|
?|ihThi|@|L?t `@| @hi hi@**) |i t|)*3i_ u@U|t @?_ @| tL*_ 4@UhLiUL?L4Ut Mi |h)?}
|L i T*@?q TThLTh@|i 4i@thit Lu UL4Li4i?| Mi|ii? |4i tihit ThLUittit tL*_
Mi _ii*LTi_ |L ThL_i @ 4i@??}u* @?tih |L |t ^it|L? Ai ?uLh4@* _tUttL?
L? UL4Li4i?|t t@**) hiuiht |L tL4i|?} U*Lti |L @ ?L|L? Lu ULhhi*@|L? OLiihc
|i |h@_|L?@* @) | U |i |4i tihit *|ih@|hi @t _i@*| | 4i@thi4i?| Lu
UL4Li4i?|t t M@ti_ L? @ ?L|L? Lu h@?! hi_U|L?
E
tii ? @?_ +i?ti*c bHH U @t
@ _gihi?| 4i@??} W? |t U@|i}Lh) Mi*L?}t |i _i@ Lu UL?|i}h@|L?
E
,?}*i @?_ Bh@?}ihc
bH.G |L ThLUittit @hi UL?|i}h@|i_ u |i tTiU|h@* _i?t|) @| uhi^i?U) 3ihL @t h@?!
L?ic UL_iTi?_i?Ui
E
BLhihL @?_ i@Ui**ic bb2c U hiuiht |L *?i@h UL4M?@|L?t Lu
ULhhi*@|i_ ThLUittit U @hi Lu *Lih @|Lhi}hitti Lh_ih |@? L|ihtc UL44L? ui@|hit
E,?}*i @?_ kL3U!c bbc i *?i@h UL4M?@|L?t U @hi ?Thi_U|@M*i | hitTiU| |L
T@t| ?uLh4@|L? @?_ UL44L? U)U*it
E
V@_ @?_ ,?}*i bb U @hi _i?i_ @t UL44L?
ui@|hit ? ht| _gihi?Uit uLh ThLUittit U @hi UL?|i}h@|i_ At U*@tt Lu UL?UiT|t
Thiti?|t tiih@* ThLM*i4t 6ht|c } UhLttULhhi*@|L? ?i|ih 4T*it ?Lh t 4T*i_ M)
UL?|i}h@|L?c UL44L? U)U*it Lh UL44L? ui@|hit
E
"@c bb @?_ 6Lh? @?_ +iU*?c
bbb 5iUL?_c |iti 4i@thit @hi M?@h) 6Lh i @4T*ic |L ThLUittit @hi i|ih UL?|i}h@|i_
Lh ?L|c M| i U@?<| it|@M*t _gihi?| _i}hiit Lu @ttLU@|L? 6?@**)c ? Lh_ih |L |it| uLh
h@?! hi_U|L?c i ?ii_ |L it|4@|i |i T@h@4i|iht Lu @ V+c U 4@) Mi ThLM*i4@|U
4
Wklv uhvhdufk kdv ehhq vxssruwhg e| dq D1U1F1 frqwudfw ri wkh Frppxqdxwh iudqfdlvh gh Ehojltxh dqg
e| wkh Hxurshdq Frpplvvlrq xqghu wkh Wudlqlqj dqg Prelolw| ri Uhvhdufkhuv Surjudpph
+
Frqwudfw Qr
HUEIPU[FW<;03546
,

i? |i ?4Mih Lu |4i tihit t *@h}i 6Lh @** |iti hi@tL?tc *i |i ?L|L? Lu h@?!
hi_U|L? t Uih|@?*) ?|ihit|?} |L U@h@U|ih3i tL4i @tTiU|t Lu |i _)?@4U ThLTih|it Lu
4*|@h@|i |4i tihitc | t ?L| |i @TThLTh@|i L?i uLh |i t|_) Lu UL4Li4i?|t
| @| t @? @TThLTh@|i 4i@thi Lu UL4Li4i?|q W? |t T@Tih i tL |@| i t|
?} |i |MLL! ^@?||it *!i ULihi?Ui @?_ ULihi?U)c U @hi _i*) ti_ ? |4i tihit
*|ih@|hic @hi ?L| @TThLTh@|i @t UL4Li4i?| ?_i it `i ThLTLti @ hi*@|i_ M| _gihi?|
4i@thic _)?@4U ULhhi*@|L?c U @hti ^|i ?@|h@**) uhL4 M@tU uhi^i?U) _L4@? ?L
|L?t #)?@4U ULhhi*@|L? U@? Mi _iUL4TLti_ M) uhi^i?U) @?_ uhi^i?U) M@?_ @?_ U@?
|i? Mi ti_ |L t|_) Mt?itt U)U*i @t i** @t *L?}h? ^it|L?t #)?@4U ULhhi*@|L?
Mi|ii? |L ThLUittit Lih @ M@?_ |h?t L| |L Mi _i?|U@* |L t|@|U ULhhi*@|L? Lu |i t@4i
ThLUittitc @u|ih t|@M*i Thi*|ih?} LhiLihc *L?}h? _)?@4U ULhhi*@|L? t hi*@|i_ ? @
t4T*i @) |L t|LU@t|U UL?|i}h@|L?
`i ti Lh ?L|L? Lu _)?@4U ULhhi*@|L? |L UL?t|hU| @ 4*|@h@|i ?_i Lu UL4Li
4i?|c U i ?@4i ULitL? Ai *@||ih ThL_it @ t44@h) 4i@thi Lu |i _i}hii Lu
UL4Li4i?| |? @ }hLT Lu @h@M*it Lh Mi|ii? |L }hLTt Lu @h@M*it @?_ U@? Mi ti_c
uLh ?t|@?Uic @t @ 4i|hU |L UL?t|hU| _)?@4U U*t|iht
AL **t|h@|i Lh ThLTLti_ 4i@thi @?_ |L ThL_i uh|ih 4L|@|L?c i it|4@|i ULi
tL? Lu L|T| _@|@ ? N5 t|@|it @?_ ,hLTi@? ?@|L?t @?_ t|_) |i uL**L?} ^it|L?t
hi L|T| U|@|L?t |? ,hLTi 4Lhi ULhhi*@|i_ |@? L|T| U|@|L?t |? |i
N5 @?_ @hi hit*|t |i t@4i uLh Mt?itt U)U*i uhi^i?Uit @?_ |i *L?} h?q hi t|@|it Lh
UL?|hit U UL4Li 4Lhi t|hL?}*) U*Lti*) *LU@|i_ uhL4 @ }iL}h@TU@* TL?| Lu iq
#Lit MLh_ih 4@||iht uLh L|T| t)?UhL?3@|L? @?_ @| @| uhi^i?U) h@?}iq
2 #)?@4U ULhhi*@|L?
2 Ai _i?|L? @?_ |i M@tU 4L|@|L?
L?t_ih |L 3ihL4i@? hi@* t|LU@t|U ThLUittit % @?_ +wi|7
%
E
b@?_7
+
E
bc Z bZc
Mi |i tTiU|h@* _i?t|) u?U|L?t Lu % @?_ + @?_
%+
E
b Mi |i ULtTiU|h4 Ai 4i@thi
i ThLTLtic _)?@4U ULhhi*@|L?c t
4
%+
E
b'
%+
E
b
t
7
%
E
b7
+
E
b
c
E

2

AL 4L|@|i |t 4i@thi i ?|hL_Ui |i tTiU|h@* _iUL4TLt|L? Lu |i ThLUittit %
|
@?_
+
|
ci
%
|
'
]
Z
3
Z
e
b|
_~
%
Eb +
|
'
]
Z
3
Z
e
b|
_~
+
Ebc E2
ihi _~
%
Eb@?__~
+
Eb @hi EUL4T*i  Lh|L}L?@* ?Uhi4i?| ThLUittit Etii i} hLU!i**
@?_ #@tc @T e , ThittL? 2 t@)t |@| %
|
@?_ +
|
U@? Mi i Thitti_ @t ??|i t4t< Lu
@it Lu _gihi?| uhi^i?Uitc i@U @?} @ h@?_L4 @4T*|_i t t i** !?L?c |i tTiU
|h@* @?_ UhLtttTiU|h@* _i?t|) u?U|L?t Lu %
|
@?_ +
|
@hi hi*@|i_ |L |i @MLi hiThiti?|@|L?
? |i uL**L?} @)G
7
%
Eb'@hEe
b|
_~
%
Eb ' @hE_~
%
Eb
7
+
E
b'@h
E
_~
%
E
b
7
%+
E
b'UL
E
_~
%
E
bc_~
+
E
b
7
+%
E
b'UL
E
_~
+
E
bc_~
%
E
b
E

`*i |i @it @TTi@h?} ? hiThiti?|@|L?
E
2 @hi UL4T*i c | t i@t*) tii? |@|c u %
|
t hi@*c |i? _~
%
E
b'_~
%
E
bc tL |@|
e
b|
_~
%
E
bne
3
b|
_~
%
E
b'2ULt
E
b|_L
%
E
b 2t?
E
b|_T
%
E
bc
E
e
ihi _L
%
@?_ _T
%
_i?L|ic hitTiU|i*)c |i hi@* @?_ |i 4@}?@h) T@h| Lu _~
%
Oi?Uii
@i |i @*|ih?@|i hiThiti?|@|L?
%
|
'2
]
Z
f
ULtEb|_L
%
Eb 2
]
Z
f
t?Eb|_T
%
Ebc ED
ihi |i UL4TL?i?| @| uhi^i?U) b t ULt
E
b|_L
%
E
bt?
E
b|_T
%
E
bc U t hi@* 54*@h*)c
|i hi@* @i _iUL4TLt|L? Lu +
|
t
+
|
'2
]
Z
f
ULt
E
b|_L
+
E
b 2
]
Z
f
t?
E
b|_T
+
E
b
t t i@t*) LM|@?i_ uhL4
E
 @?_
E
ec Lh ThLTLti_ 4i@thic 4
%+
E
bc t ?L|?} i*ti |@?
|i ULhhi*@|L? ULiUi?| Mi|ii? |i hi@* @it Lu uhi^i?U) b @TTi@h?} ? |i @MLi
hiThiti?|@|L?c i ULt
E
b|_L
+
E
b t?
E
b|_T
+
E
b@?_ULt
E
b|_L
%
E
b t?
E
b|_T
%
E
b At
t |i hi@tL? ) |i 4i@thi t _i?i_ L?*) uLh ?L??i}@|i uhi^i?Uitc L? |i ?|ih@*
d
fcZ *i@h*)c _)?@4U ULhhi*@|L? U@? @h) Mi|ii? @?_ 

22 Ai hi*@|L? | RULihi?U) @?_ RULihi?Ui
#)?@4U ULhhi*@|L? t t|hU|*) hi*@|i_ |L ML| ULihi?U) @?_ ULihi?UiQ|L i** !?L?
?_iUit ? |4i tihit *|ih@|hi Lihi?U) t _i?i_ @t
F
%+
Eb'
7
%+
Eb
t
7
%
Eb7
+
Eb
'
%+
Ebn'
%+
Eb
t
7
%
Eb7
+
Eb
c
ihi '
%+
Eb |i ^@_h@|hi tTiU|h4 Oi?Ui ULihi?U) t UL4T*i ? }i?ih@* @?_ t ?L|
t)44i|hUc i F
%+
Eb@?_F
+%
Eb @hi ?L| i^@*c M| UL?}@|i #)?@4U ULhhi*@|L? 4
%+
Eb
t |i hi@* T@h| Lu ULihi?U) @?_ U@? @*tL Mi LM|@?i_ M) @ih@}?} ULihi?Uit @| uhi^i?U)
b @?_ bc U @*tL @hi UL?}@|iG
4
%+
E
b'
F
%+
E
bnF
%+
E
b
2
)t?}
E
 |i hi@_ih U@? i@t*) ihu) |@| F
%+
E
b t |i ULhhi*@|L? ULiUi?| Lu _~
%
E
b
@?_ _~
+
E
b Oi?Ui |i ?|ihThi|@|L? Lu ULihi?U) t ^|i t4*@h |L |@| Lu _)?@4U
ULhhi*@|L?c |i M@tU _gihi?Ui Mi?} |@| |i uLh4ih t hi*@|i_ |L |i UL4T*i hiThiti?|@|L?
E
2 h@|ih |@? |i hi@* @i _iUL4TLt|L?
E
D | @?L|ih @)c i t4T*) UL**iU| |L}i|ih
uhi^i?Uit b @?_ b tL |@| |i 4@}?@h) T@h|t U@?Ui* L| At t TihuiU|*) UL?tt|i?|
| Lh @4c t?Ui |i @it Lu uhi^i?Uit b @?_ b @i|it@4iTihL_U|)
+i}@h_?} ULihi?UiQLh t^@hi_ULihi?U)c @t t tL4i|4it U@**i_Q| t _i?i_ @t |i
t^@hi_ 4L_*t Lu ULihi?U)c |@| t
M
%+
E
b'
%+
E
b
2
n '
%+
E
b
2
7
%
Eb7
+
Eb
'
m 7
%+
E
b m
2
7
%
Eb7
+
Eb
AihiuLhi ULihi?Ui t hi@* @?_ t)44i|hU OLiihc | _Lit ?L| 4i@thi ULhhi*@|L? @|
_gihi?| uhi^i?Uitc MiU@ti | _thi}@h_t |i T@ti _gihi?Uit Mi|ii? @h@M*it At U@?
Mi i@t*) tii? M) LMtih?} |@| M
%+
E
b t ?@h@?| | hitTiU| |L tu|?} |i ThLUittit
Lih |4ic i |i ULihi?Ui Mi|ii? %
|
@?_ +
|
3
&
t|it@4i@t|@|Lu%
|
@?_ +
|
Atc
Mit_it Mi?} @? 44i_@|i UL?ti^i?Ui Lu |i _i?|L?c t UL?tt|i?| | |i i**!?L?
?|ihThi|@|L? Lu ULihi?Ui @t |i -
2
uhL4 |i hi}hittL? Lu +
|
L? |i T@t|c Thiti?| @?_ u|hi
Lu %
|
G U*i@h*)c ? |t hi}hittL?c i|ih |i UL?|i4TLh@?iLt hi}hittLh t %
|
Lh %
|
3
&
U@??L|
4@!i @?) _gihi?Ui
_gihi?| @) |L tii |i t@4i TL?| t ThL_i_ M) |i TL*@h hiThiti?|@|L? Lu |i
UhLtttTiU|h4G
7
%+
E
b'-
%+
E
be
w
E
b
e

Citations
More filters
Dissertation
31 Oct 2013

1 citations


Cites background from "A Measure of Comovement for Economi..."

  • ...As proposed by Croux et al. (2001), we calculate the dynamic correlation between each country pair which shows the degree of synchronization at a given frequency....

    [...]

Posted Content
TL;DR: In this article, the authors examined the co-movement between a leading first-world economy (Germany) and an emerging market economy (South Africa) by applying a dynamic factor model.
Abstract: This article examines the co-movement between a leading first-world economy (Germany) and an emerging market economy (South Africa) by applying a dynamic factor model. These countries have been chosen as proxies to analyse the channels of transmission of positive supply and demand shocks in developed economies and the effects of these on emerging market economies. The study concludes that supply and demand shocks in developed economies do not necessarily have similar effects in emerging market economies. A German supply shock has more of a demand-shock effect on the South African economy, while a German demand shock has more of a monetary policy effect on the South African economy. This implies that the policy response in emerging market economies should not necessarily be the same as in developed economies. In the case of the transmission of a positive supply shock from a developed country to an emerging economy, the demand effect will lead to increase in prices, which will require a more restrictive monetary policy stance. Similarly, a positive demand shock from a developed economy is transmitted as a monetary policy shock in an emerging market economy, requiring the latter group of countries to stimulate demand through expansionary fiscal and/or monetary policy.

1 citations

Journal ArticleDOI
TL;DR: In this paper , the authors investigated the interdependence between economic policy uncertainty and business cycles within and among six emerging market economies (EMEs) from January 1999 to December 2018.
Abstract: Purpose The purpose of this paper to investigate the interdependence between economic policy uncertainty (EPU) and business cycles within and among six emerging market economies (EMEs) from January 1999 to December 2018. Design/methodology/approach This study adopts the wavelet multiple correlations and wavelet multiple cross-correlation (WMCC) based on the maximal overlap discrete transform estimator. This methodology simultaneously investigates how two or more time series variables move together continuously at both time and frequency domains. Findings The empirical results show that business cycles comove with EPU for both intra- and inter-country analysis, with the long term showing the greatest degree of interdependence. In intra-country comparisons, EPU has a positive correlation with consumer price index and a negative correlation with share price index. According to the WMCC results, EPU does not have any leading or lagging power within each EME, but rather import has both lead and lag power. The inter-country WMCC results are all significant, with Korea’s EPU leading/following all EMEs across all scales. Originality/value This study contributes to the ongoing debate about what causes business cycles to comove by investigating business cycle indicators (leader/follower) using a robust wavelet methodology. The authors propose new variables that can clearly reflect the outcome of economic policy actions and translate information about EPU shocks. The inclusion of the variables has altered the understanding of the relationship between EPU and business cycle fluctuations. Policymakers also gain new insights into the trends and patterns of EPU and business cycles, which will help them formulate and implement fiscal and monetary policies more effectively.

1 citations

Book ChapterDOI
01 Jan 2020
TL;DR: The codispersion coefficient is a geometrically natural comovement coefficient in that it compares proportional slopes at matched pairs of points across sequences (Croux et al. 2001), and has been extensively used in multivariate spatial prediction.
Abstract: In this chapter, we study another measure to quantify the assessment of two spatial or temporal series. This coefficient, called the codispersion coefficient, was first introduced by Matheron (1965) and has been used in several applications (Goovaerts 1994, 1997, 1998; Chiles and Delfiner 1999; Blanco-Moreno et al. 2005; Vallejos 2008, 2012; Buckley et al. 2016a, b). It is a normalized version of the cross-variogram of two spatial sequences and has been extensively used in multivariate spatial prediction (Ver Hoef and Barry 1998). Rukhin and Vallejos (2008) studied the codispersion coefficient from both theoretical and applied perspectives. This coefficient has also been studied in time series to address how two time sequences change concurrently; the codispersion coefficient is a geometrically natural comovement coefficient in that it compares proportional slopes at matched pairs of points across sequences (Croux et al. 2001). Applications and extensions of the codispersion coefficient can be found in Vallejos et al. (2015).

1 citations

References
More filters
Journal ArticleDOI
TL;DR: The relationship between co-integration and error correction models, first suggested in Granger (1981), is here extended and used to develop estimation procedures, tests, and empirical examples.
Abstract: The relationship between co-integration and error correction models, first suggested in Granger (1981), is here extended and used to develop estimation procedures, tests, and empirical examples. If each element of a vector of time series x first achieves stationarity after differencing, but a linear combination a'x is already stationary, the time series x are said to be co-integrated with co-integrating vector a. There may be several such co-integrating vectors so that a becomes a matrix. Interpreting a'x,= 0 as a long run equilibrium, co-integration implies that deviations from equilibrium are stationary, with finite variance, even though the series themselves are nonstationary and have infinite variance. The paper presents a representation theorem based on Granger (1983), which connects the moving average, autoregressive, and error correction representations for co-integrated systems. A vector autoregression in differenced variables is incompatible with these representations. Estimation of these models is discussed and a simple but asymptotically efficient two-step estimator is proposed. Testing for co-integration combines the problems of unit root tests and tests with parameters unidentified under the null. Seven statistics are formulated and analyzed. The critical values of these statistics are calculated based on a Monte Carlo simulation. Using these critical values, the power properties of the tests are examined and one test procedure is recommended for application. In a series of examples it is found that consumption and income are co-integrated, wages and prices are not, short and long interest rates are, and nominal GNP is co-integrated with M2, but not M1, M3, or aggregate liquid assets.

27,170 citations

01 Jan 1987

3,983 citations


"A Measure of Comovement for Economi..." refers background in this paper

  • ...In this category belong the following three concepts: (i) the idea of co-integration (Engle & Granger, 1987): two processes are co-integrated if the spectral density at frequency zero has rank one; (ii) codependence (Gourieroux & Peaucelle, 1992), which refers to linear combinations of correlated…...

    [...]

Journal ArticleDOI
TL;DR: In this article, the authors present evidence that most of the unemployment fluctuations of the seventies (unlike those in the sixties) were induced by unusual structural shifts within the U.S. economy.
Abstract: A substantial fraction of cyclical unemployment is better characterized as fluctuations of the "frictional" or "natural" rate than as deviations from some relatively stable natural rate. Shifts of employment demand between sectors of the economy necessitate continuous labor reallocation. Since it takes time for workers to find new jobs, some unemployment is unavoidable. This paper presents evidence that most of the unemployment fluctuations of the seventies (unlike those in the sixties) were induced by unusual structural shifts within the U.S. economy. Simple time-series models of layoffs and unemployment are constructed that include a measure of structural shifts within the labor market. These models are estimated and a derived natural rate series is constructed.

1,128 citations

ReportDOI
TL;DR: In this paper, the authors introduce a class of statistical tests for the hypothesis that some feature that is present in each of several variables is common to them, which are data properties such as serial correlation, trends, seasonality, heteroscedasticity, auto-regression, and excess kurtosis.
Abstract: This article introduces a class of statistical tests for the hypothesis that some feature that is present in each of several variables is common to them. Features are data properties such as serial correlation, trends, seasonality, heteroscedasticity, autoregressive conditional hetero-scedasticity, and excess kurtosis. A feature is detected by a hypothesis test taking no feature as the null, and a common feature is detected by a test that finds linear combinations of variables with no feature. Often, an exact asymptotic critical value can be obtained that is simply a test of overidentifying restrictions in an instrumental variable regression. This article tests for a common international business cycle.

550 citations

Posted Content
TL;DR: The existence of a serial correlation common feature among the first differences of a set of I(1) variables implies the existence of common cycle in the Beveridge-Nelson-Stock-Watson decomposition of those variables as mentioned in this paper.
Abstract: The existence of a serial correlation common feature among the first differences of a set of I(1) variables implies the existence of a common cycle in the Beveridge-Nelson-Stock-Watson decomposition of those variables. A test for the existence of common cycles among cointegrated variables is developed. The test is used to examine the validity of the common trend-common cycle structure implied by Flavin's excess sensitivity hypothesis and Campbell and Mankiw's mixture of rational expectations and rule-of-thumb hypothesis for consumption and income. Linear independence between the cointegration and the cofeature vectors is exploited to decompose consumption and income into their trend and cycle components. Copyright 1993 by John Wiley & Sons, Ltd.

511 citations