Wavelet-based Relationships between Stock Returns and Industrial Production

• 발행기관: 한국자료분석학회
• 저자명: 이연정(Yeonjeong Lee) 강주화(Zhuhua Jiang) 윤성민(Seong-Min Yoon)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.17 No.3, 1203~1215쪽, 전체 13쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2015.06.30

영문 초록

Using the maximum overlap discrete wavelet transform (MODWT), this study investigates relationships between real stock returns and real industrial production growth for Korea, Japan, the US, and the UK, during the period from August, 1992 to December, 2013. This study found, through wavelet-based Granger causality analysis, that relationships between variables not found in the raw data could be estimated on a scale-by-scale basis. Second, by separating the long-term feedback relationships between variables from the original data, the series derived from the wavelet decomposition provides useful information for market participants who are interested in high-frequency movements of variables. Finally, when the wavelet filter is adapted sequentially, it is useful to know the net movement of high-frequency variables and the degree of common movement between variables reflecting the impact of low-frequency market shocks.

목차

1. Introduction

2. Methodology

3. Empirical results

4. Conclusions

References

키워드

MODWT Granger causality Stock returns Industrial production index