Price discovery in Taiwan futures markets

• Main Authors: Ting-Lin Liu, 劉廷麟
• Other Authors: Wen-Liang Hsieh
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/49178049453284441910

碩士 === 淡江大學 === 財務金融學系 === 89 === Title of Thesis: Price discovery in Taiwan futures markets Total Pages: 94 Name of Institute: Graduate Institute of Money, Banking and Finance, Tamkang University Graduate Date: June, 2001 Degree Conferred: Master Name of Student: Ting-Lin Liu Advisor: Wen-liang Hsieh ABSTRACT Price discovery is the process by which markets attempt to find equilibrium prices. This thesis compares the price discovery function of two similar index futures: the TAIFEX traded Taiwan Stock Exchange Capitalization Weighted Stock Index futures and the SGX traded Morgan Stanley Capital International Taiwan Index futures. We are interested in the speed of price adjustment of two futures as well as the underlying spot indices. Using high frequency and synchronized transaction data, we investigate contribution of price discovery within each of the 5 grouped markets. Results indicate that the price series in each of five groups are a cointegrated system. Estimated coefficients of the vector error correction model suggest that: 1.In category 1 (TAIFEX spot and TAIFEX futures), TAIFEX future serves the dominant price discovey function. 2.In category 2 (SGX spot and SGX futures), SGX future and its underlying spot price discovey function are about the same. 3.In category 3 (TAIFEX futures and SGX futures), price adjustment takes place in SGX future faster than TAIFEX future. 4.In category 4 (TAIFEX spot and SGX spot), SGX spot serves the dominant price discovery function. 5.In category 5 (TAIFEX spot, TAIFEX futures, SGX spot and SGX futures), four series show equal ability in price discovery. Conclusions are as follows: First, evidences support that trading cost and infrequent trading are important factors of the price discovery. Next, data frequency and synchronicity will affect the results of error correction model. It is optimal to use trade-by-trade data instead of 5-minute data. Third, the price discovery function of futures market is affected by trading mechanism and market maturity.