nowcasting of gross regional product and analyzing regional business cycle nariyasu yamasawa

36
Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Upload: miranda-newman

Post on 21-Jan-2016

215 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle

Nariyasu Yamasawa

Page 2: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Introduction

• Which data and method is relevant to measure Regional business cycle?

• How similar are the prefectures business cycle? • To what extent have prefecture' recession and

expansion experiences been in sync each other ?

• what might explain the differences in business cycle ?

Page 3: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

• 1.Data• 2.Methodology• 3.Emprical Result• 4.Conclusion

My presentation has 4 parts.

Page 4: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

1.Data Problem in Japan

Annual Quarterly, MonthlyNational GDP Composite Index,

GDPRegional GRP

2years lagComposite Index, Industrial productionMonthly GRP

GDP and GRP(Gross Regional Product) is the best for analyzing business cycle.

For the analysis of the regional business cycle, the key is the choice of the data which represent the business cycle.

Page 5: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Nowcasting of GRP

Present

Official GRP

TIME

Real Monthly GRP

Past

2 years lag

90 days Lag

Page 6: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

• The Japanese Cabinet Office has been releasing data on most of the GRP components in the form of a monthly index called the Regional Domestic Expenditure Index (RDEI) since May 2012,.

• We estimated the rest of components, that is Government Consumption, Net export.

• By summing up the expenditure components, we could estimate the monthly GRP.

How to estimate monthly GRP?

Page 7: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Item Methodology for estimation

Private consumption Divided by 44 types of consumption

Private residential investment

“Statistics of Construction Starts of Residential Properties”

Private fixed investment

Estimated by building, construction, machinery, aircraft, motor vehicles, and other transportation machinery

Public investment “Statistics of Construction Order by 47 Prefectures”

Government Consumption Estimated by Author.

Net Export Estimated by Author

RDEI

Page 8: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Real Monthly GRP for 47 prefectures

(Note) The data Start from April 2002

Page 9: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

2.MethodologyHow do we measure business cycle?

• Band Pass filter (Baxter and King Filter)– Can remove noise and trend– Considered Business Cycle Frequency 18months – 96 months– Weak point : Baxter King filter cannot analyze present situation

• Regime Switching model(Hamilton(1989)) – Suppose mean growth rate switches between high- and low-

growth regimes.– Probability of recession – Apply spatial analysis with weights matrix W

We tried to extract business cycle by two types of methodology.

Page 10: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Band Pass Filter

ntntttntnt xBxBxBxBxBy 11011

18

2,

96

2,

1,)sin()sin(

0

baab

B

nn

nanbBn

Page 11: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Band Pass Filter

Level data After filtering

Page 12: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Regime Switching Model

We use regime switching model which enable us to identify recession period and expansion period. It suppose the series can divided by two regimes, that is, the high growth rate and low growth rate regime.

Page 13: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Regime Switching model

Growth Rate Recession Probability

Page 14: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Spatial Model

Page 15: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Spatial Contiguity Weights Matrix• Indicate whether prefectures share a

boundary or not. • Neighbors = 1

Then, calculate the share.

Page 16: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

3. Estimation Result

• Analyze Mainly Great Recession 2008-2009

Page 17: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Result (Band Pass Filter)

Tokyo

Page 18: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Recession Period(Filter)123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748

20142003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

PREFECTURES

Page 19: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Result(Band Pass filter)

• Almost all the data have the same behavior.• Tokyo is not the same. The peak is 10 months

earlier than the official peak(Feb. 2008).• Hokkaido, Aomori, Kanagawa went in

recession earlier.• Osaka, Hyogo, Aichi went in recession later.

Page 20: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Result of Regime Switching Model

GDPMGGDPATGDPYGGDPFSGDPIGGDPTGGDPGM

GDPSTGDPCBGDPTKGDPKGGDPNGGDPTYGDPIKGDPFIGDPYN

GDPNNGDPGFGDPSOGDPACGDPMEGDPSIGDPKTGDPOS

GDPHGGDPNRGDPWYGDPTTGDPSNGDPOYGDPHSGDPYI

GDPTSGDPKWGDPEMGDPKCGDPFOGDPSGGDPNSGDPKMGDPOTGDPMZGDPKSGDPON

0.00.10.20.30.40.50.60.70.80.91.0

2002

/04

2002

/12

2003

/08

2004

/04

2004

/12

2005

/08

2006

/04

2006

/12

2007

/08

2008

/04

2008

/12

2009

/08

2010

/04

2010

/12

2011

/08

2012

/04

2012

/12

2013

/08

2014

/04

GDPHK

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPAO

GDPAO

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPIW

GDPIW

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPMG

GDPMG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPAT

GDPAT

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPYG

GDPYG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPFS

GDPFS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPIG

GDPIG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPTG

GDPTG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPGM

GDPGM

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPST

GDPST

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPCB

GDPCB

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPTK

GDPTK

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKG

GDPKG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPNG

GDPNG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPTY

GDPTY

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPIK

GDPIK

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPFI

GDPFI

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPYN

GDPYN

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPNN

GDPNN

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPGF

GDPGF

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPSO

GDPSO

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPAC

GDPAC

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPME

GDPME

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPSI

GDPSI

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKT

GDPKT

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPOS

GDPOS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPHG

GDPHG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPNR

GDPNR

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPWY

GDPWY

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPTT

GDPTT

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPSN

GDPSN

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPOY

GDPOY

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPHS

GDPHS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPYI

GDPYI

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPTS

GDPTS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKW

GDPKW

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPEM

GDPEM

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKC

GDPKC

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPFO

GDPFO

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPSG

GDPSG

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPNS

GDPNS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKM

GDPKM

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPOT

GDPOT

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPMZ

GDPMZ

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPKS

GDPKS

0.00000

0.20000

0.40000

0.60000

0.80000

1.00000

1.20000

2002

/04

2003

/04

2004

/04

2005

/04

2006

/04

2007

/04

2008

/04

2009

/04

2010

/04

2011

/04

2012

/04

2013

/04

2014

/04

GDPON

GDPON

Page 21: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Recession Period(Regime Switching)123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748

2003 2004 2005 2006 2007 2013 20142008 2009 2010 2011 2012

PREFECTURES

Page 22: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Recession(Regime Switching)

0 400kmAug- 08

( )景気後退期

0. 8

0 400kmSep- 08

( )景気後退期

0. 8

0 400kmOct - 08

( )景気後退期

0. 8

• Prefectures in blue are in recession.• Hokkaido, Kagoshima, Yamaguchi, Hiroshima,Niigata, Shizuoka is earlier.• Prefectures which are far from Tokyo tend to go recession early.

Tokyo

Page 23: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

It took 4 months for almost all prefectures to be in recession.

Page 24: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Spatial Model

Page 25: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Estimation of ρ

Prefecturesρ  is positive and significant.

Page 26: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

What might explain the differences in business cycle ?

• Dependent Variables = Beginning date of Recession– Date is number of days from 1900/1/1 – e.g.Feb.2008 is “39479”

• Explanation Variables various kind of data

Cross Section Regression ,Year of 2008

Page 27: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Explanation variablesPOP Population(person)

OLD Population of Old age(%of Total)

PERCAPITA Gross Regional Income Per capita(Yen)

HIGH High School Enrolment(%)

UNIV University Students(% of Total Population)

GDPAGRI Agriculture (% of GRP)

GDPMANU Manufacturing(% of GRP)

GDPCONST Construction(% of GRP)

GDPGOV Government Service(% of GRP)

RIPUB Public Construction(% of GRP)

LEND Lending(% of GRP)

TK Tokyo Dummy

KG Kanagawa Dummy

Page 28: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Regression Result

Peak Date(Fileter) Peak Date(Markov)Variable Coefficient S.E. Coefficient S.E.

Constant 39466 149.38 *** 40307 307.82 ***Population(person) 0.00 0.00 0.00 0.00Population of Old age(%of Total) 1.80 2.05 -9.44 3.75 **Gross Regional Income Per capita(Yen) -0.08 0.05 * -0.21 0.12 *High School Enrolment(%) 1.13 0.93 6.51 3.25 *University Students(% of Total Pop) -2.14 6.12 -28.73 20.02Agriculture (% of GRP) -2.47 4.53 -5.97 12.73Manufacturing(% of GRP) 2.30 1.02 ** 0.29 2.77Construction(% of GRP) 5.89 3.31 * -1.83 7.56Government Service(% of GRP) -1.13 3.11 1.56 10.12Public Construction(% of GRP) -0.60 4.91 -15.87 15.85Lending(% of GRP) 66.85 47.76 -14.34 83.38Tokyo Dummy 1061.08 102.46 *** 981.34 225.87 ***Kanagawa Dummy -43.27 33.82 -202.25 98.12 **

Adj R2 0.959 0.555

HAC standard errors & covarianceStatistically Signifcant at the 10% level(*),5%level(**),1%level(***).Included observations: 47

Page 29: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

What affects the business cycle?Frequency Filter Markov Switching

Model1 Share of Manufacturing

Industry(%of GRP)Population Share of Old age (%)

2 GRP Per Capita(Yen) GRP per Capita(yen)

3 Share of Construction Industry(% of GRP)

High school enrollment (%)

Page 30: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Conclusion

• Which data is relevant to measure Regional business cycle. Monthly GRP

• How similar are the prefectures business cycle? Recession date is different, the range is about 4months

• To what extent have prefecture' recession and expansion experiences been in sync with each other's ? Neighborhood Effect exists

• what might explain the differences in business cycle ? GRP per capita etc.

Page 31: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Reference• Michael T. Owyang, Jeremy Piger and Howard J. Wall Source(2005)” Business Cycle

Phases in U.S. States” The Review of Economics and Statistics,Vol. 87, No. 4 (Nov., 2005), pp. 604-616

• Hamilton, J. D., (1989)"A New Approach to the Economic Analysis of Nonsta- tionary Time Series and the Business Cycle," Econometrica 57:2 (1989), 357-384.

Page 32: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

0 400kml ength

( )累積表示

987654321

Earlier prefecture is darker

Page 33: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

0 400km9- J ul

( )景気後退期

0. 8

Recovery (Regime Switching)

0 400km9- Mar

( )景気後退期

0. 8

0 400km9- May

( )景気後退期

0. 8

• Prefectures in Red are in recession.• Kagoshima, Yamaguchi, Osaka, Saitama, Chiba recovered earlier.• Prefectures which are near Tokyo and Osaka recovered earlier.

Page 34: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Spatial Weight Matrix• Based on Distance– Nearest Neighbor Weights– Radial Distance Weights– Power Distance Weights– Exponential Distance Weights

• Based on Boundaries– Spatial Contiguity Weights– Shared-Boundary Weights

Stakhovych and Bijmolt (2009)

Page 35: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Band Pass Filter

Regime Switching Model

Data Level Growth ratePeak Max End point of High

growth regime

Bottom Min End point of Low growth regime

Data Limitation

Need 18month lag and lead

Influenced outlier

Page 36: Nowcasting of Gross Regional Product and Analyzing Regional Business Cycle Nariyasu Yamasawa

Regime Switching Model

Markov Chain : Any persistence in the regime is completely summarized by the value of the state in the last period.