An Investigation on the Price and Volume Relationship of Taiwan Interest Rate Futures

• Main Authors: Cheng-hsiang Wen, 溫呈祥
• Other Authors: Chui-chun Tsai
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/ah5r2r

碩士 === 靜宜大學 === 會計學系研究所 === 94 === This paper uses VAR-bi-EGARCH model to investigate the lead/lag relationship and asymmetric volatility between return and trading volume of nearby-month 10-year government bond futures(GBF) and 30-day Commercial Paper futures(CPF). The results are as follows：(1) According to Spearman’s rank correlation coefficient and VAR to test the contemporaneous correlations between return and trading volume of both GBF and CPF are not significantly. The demonstrations contradicts the Mixture of Distribution Hypothesis and the Sequential Information Arrival Model. (2) According to bi-EGARCH Ⅰ、The half life of return volatility of GBF is 0.66 day. With the same method, the half life of return volatility of CPF is 0.7 day. Ⅱ、The bad news occurred in lagged return volatility of CPF tend to enlarge volatility of itself. Ⅲ、Large unexpected shocks in return volatility, trading volume volatility of GBF and trading volume volatility of CPF induce higher volatility in itself. (3) With the same method Ⅰ、In GBF, good news occurred in lagged trading volume volatility tend to enlarge volatility of return. In CPF, bad news occurred in lagged return volatility tend to enlarge volatility of trading volume, and vice versa. Ⅱ、Small unexpected shocks in trading volume volatility of GBF induce higher volatility in return. But small unexpected shocks in return volatility of CPF induce higher volatility in trading volume.So, in GBF, volatility of return and trading volume wouldn’t follow the interrelationship of volatility effects, but trading volume volatility significant lead return volatility. Therefore, bad news occurred in trading volume volatility of CPF will cause higher volatility of return. The results supports the Sequential Information Arrival Model that return volatility is potentially forecastable with knowledge of trading volume.