臺灣博碩士論文加值系統

本文主要是探討變數間的變異數因果關係，本文採用義大利學者Massimiliano Caporin 於2007年發表的指數因果模型Exponential Causality multivariate GARCH (EC-GARCH) model。此模型結構由一個指數函數乘上傳統GARCH模型而成，使用指數函數可以將變數間的影響直接的表達出來。有鑑於中國的股票市場在國際間愈來愈重要，本論文主要的四個變數選用上海A指、深圳A指兩指數其報酬率波動及成交量變動。

各變數資料期間為1996年1月5日到2012年4月6日的週資料，818個觀察值。藉由檢驗變數間的變異數因果關係，了解變數間影響方向及影響程度，以提供交易者有利的資訊。在檢測四個模型之後，本研究結果發現四個模型皆得到同樣的結果，亦即本期報酬率波動會影響下一期成交量的變動。由此可了解，上海A指及深圳A指，其報酬率的波動扮演較重要的角色，交易者藉由觀察報酬率波動，可了解到成交量的波動情形。

This study investigates on the variance causality by applying the Exponential Causality multivariate GARCH (EC-GARCH)model, which was introduced by Massimiliano Caporin in 2007. The main structure of this model is an exponential factor multiplying the traditional GARCH equation to drive the causality relation.
This research mainly focuses on the biggest stock exchange of China, selecting return and trading volume of both Shanghai Stock Exchange A Share Index and Shenzhen Stock Exchange A Share Index as our four variables. In addition, used data is weekly data from January 5, 1996 to April 6, 2012 with total 818 observations for each variable. Our purpose is to analyze the variance causality among four variables to provide some useful information for the traders.
As the result, by testing on the variance causality between Shanghai Stock Exchange A Share Index and Shenzhen Stock Exchange A Share Index, we find out an unexpected return volatility shock at time t-1 will induce the investors to invest on the stock at the time. The results emphasize the importance of return volatility on predicting the volume change of Shenzhen Stock Exchange A Share Index and Shanghai Stock Exchange A Share Index. The traders in the stock market may realize the volume change from investigating the return volatility.

List of Content

Chapter1 Introduction 1
1.1 Background of the research 1
1.2 Purpose of the research 4
1.3 The structure of the research 4
Chapter 2 Literature Review 6
Chapter 3 Methodology 10
3.1 Unit Root Test 10
3.2 Test for serial correlation 12
3.2.1 Box and Pierce’s test 12
3.2.2 Ljung-Box Test 13
3.3 ARCH Effect Test 13
3.4 Establish Bivariate EC-GARCH model 14
Chapter 4 Empirical Results and Analysis 18
4.1 Data Description 18
4.1.1 Data Source 18
4.1.2 Trend Plot of the data 18
4.1.3 Descriptive statistic Analysis 23
4.2.1 Unit Root Test 26
4.2.3 Autoregressive Conditional Heteroscedasticity Test 30
4.2 Empirical result 31
4.2.1 Model 1 : Analysis of variance causality between return and Trading volume
of Shanghai Stock Exchange A Share Index 31
4.2.2 Model 2 : Analysis of variance causality between return and trading volume
for Shenzhen Stock Exchange A Share Index 34
4.2.3 Model 3 : Analysis of variance causality between return of Shanghai Stock
Exchange A Share Index and Trading volume of Shenzhen Stock Exchange A Share Index 37
4.2.4 Model 4 : Analysis of variance causality between returns of Shenzhen Stock
Exchange A Share Index and Trading volume of Shanghai Stock Exchange A Share Index 40
Chapter 5 Conclusions and Suggestions 43
5.1 Conclusion 43
5.2 Suggestions for Future research 45
REFERENCES 46

1.Caporin, M. (2007). “Variance (non) causality in multivariate GARCH”. Econometric Reviews. 26(1):1-24.
2.Cheung, Y., Ng, L. K. (1996). “A causality-in-variance test and its application to financial market prices”. Journal of Econometrics 72:33–48.
3.Chiang, T. C., Qiao, Z., Wong, W. K., (2010). “New evidence on the relation between return volatility and trading volume”. Journal of Forecasting. 29, 502-515
4.Chuang, W. I., Liu, H. H., Susmel, R. (2012). “ The bivariate GARCH approach to investigating the relation between stock returns, trading volume, and return volatility”. Global Finance Journal 23:1-15.
5.Comte, F., Lieberman, O. (2000). “Second-order noncausality in multivariate GARCH processes.” Journal of Time Series Analysis 21(5):535–557.
6.Engle, R. F. (2002). Dynamic conditional correlation: a simple class of multivariate generalized autoregressive conditional heteroskedasticity models. Journal of Business and Economic Statistics 20(3):339–350.
7.Granger, C. W. J. (1969). Investigating causal relations by econometric models and cross spectral methods. Econometrica 37:424–438.
8.Granger, C. W. J. (1980). Testing for causality: a personal viewpoint. Jounrnal of Economics Dynamic and Control 2(4):329–352.
9.Granger, C. W. J. (1988). Some recent developments in a concept of causality. Journal of Econometrics 39:199–211.
10.Granger, C. W. J., Robins, R. P., Engle, R. F. (1986). Wholesale retail prices: bivariate time series modeling with forecastable error variances. In: Model reliability. Boston MA: MIT Press.
11.Gallant, A. R., Rossi, P. E., Tauchen, G. (1992). Stock price and volume. Review of Financial Studies 5:199-242.
12.Hiemstra, C., Jones, J. D. (1994). Testing for Linear and Nonlinear Granger Causality in the Stock Price-Volume Relation. The Journal of Finance
13.Hafter, C. M., Herwartz, H. (2004). “Testing for causality in variance using multivariate GARCH models”. Econometric Institute Report 20.
14.Lee, B. S., Rui, O. M. (2002). “The dynamic relationship between stock returns and trading volume: Domestic and cross-country evidence”. Journal of Banking & Finace. 26(1), 34-37.