2012
DOI: 10.5018/economics-ejournal.ja.2012-5
|View full text |Cite
|
Sign up to set email alerts
|

The Japanese Economy in Crises: A Time Series Segmentation Study

Abstract: The authors performed a comprehensive time series segmentation study on the 36 Nikkei Japanese industry indices from 1 January 1996 to 11 June 2010. From the temporal distributions of the clustered segments, we found that the Japanese economy never fully recovered from the extended 1997-2003 crisis, and responded to the most recent global financial crisis in five stages. Of these, the second and main stage affecting 21 industries lasted only 27 days, in contrast to the two-and-a-half-years acrossthe-board reco… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
22
0

Year Published

2012
2012
2019
2019

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 26 publications
(22 citation statements)
references
References 123 publications
0
22
0
Order By: Relevance
“…In order to do so, I shall use some techniques taken from Random Matrix Theory [1], first developed for the use in nuclear physics and then used in many areas, including finance (see [2] for a comprehensive list of contributions). There are many studies of networks built from data of financial markets around the world, mainly based on the New York Stock Exchange [3]- [18], but also using data from Nasdaq [18], the London Stock Exchange [19] [20], the Tokyo Stock Exchange [19] [21], the Hong Kong Stock Exchange [19], the National Stock Exchange of India [22], the Global financial market [11] [19] [23]- [26], the USA Commodity market [27], the foreign currency market [28]- [31], and the world trade market [32]- [35]. Until the present date, to the author's knowledge, no work has been done in using the stocks of BM&F-Bovespa as a source of data for developing networks.…”
Section: Introductionmentioning
confidence: 99%
“…In order to do so, I shall use some techniques taken from Random Matrix Theory [1], first developed for the use in nuclear physics and then used in many areas, including finance (see [2] for a comprehensive list of contributions). There are many studies of networks built from data of financial markets around the world, mainly based on the New York Stock Exchange [3]- [18], but also using data from Nasdaq [18], the London Stock Exchange [19] [20], the Tokyo Stock Exchange [19] [21], the Hong Kong Stock Exchange [19], the National Stock Exchange of India [22], the Global financial market [11] [19] [23]- [26], the USA Commodity market [27], the foreign currency market [28]- [31], and the world trade market [32]- [35]. Until the present date, to the author's knowledge, no work has been done in using the stocks of BM&F-Bovespa as a source of data for developing networks.…”
Section: Introductionmentioning
confidence: 99%
“…Clustering has been used to identify structure in subsequences from time series data, though [13] describes how naive segmentation can lead to meaningless clusters. Segmentation points are statistically determined in [5,15]; clustering is performed on the time series segments and models built from the segments respectively. We are using a form of subsequence clustering to determine characteristic features, where subsequences are determined through change points in the mean.…”
Section: Related Workmentioning
confidence: 99%
“…Extensive literature describes the detection of structural breaks or change points separating stationary segments; this followed the pioneering works of Goldfeld and Quandt [14]. More recently, a recursive entropic scheme to separate financial time series has been proposed [15].…”
Section: Introductionmentioning
confidence: 99%
“…The likelihood-ratio test is one of the most efficient tests of statistical hypotheses [1]. Recently, Cheong et al have considered a one-dimensional Gaussian case [15] in the context of financial time series analysis. The likelihood-ratio test is a kind of hypothesis test between a null hypothesis and an alternative model.…”
Section: Introductionmentioning
confidence: 99%