本文使用高頻日內資料探討權證與現貨市場間的價格發現關係。利用樣本期間為2007年10月至2008年10月份的現貨與權證的交易資料。上述權證的標的資產與現貨相同。本文將在確定兩個市場間是否具有共整合(cointegration)關係之後來決定本次研究該採用向量誤差修正模型(vector error correction model, VECM)或者向量自我回歸模型(VAR)，最後藉由衝擊反應函數以及變異數分解來探討市場間相互影響的程度。經由實證結果可發現，現貨的領導地位較強於權證。而且兩個市場對於對方幾乎沒有解釋能力。這可能與台股的權證市場尚未發展成熟有關。雖然低交易成本的價格發現能力越強，但交易量也是影響的重要因素之一。 This thesis aims to investigate the price discovery between stocks and warrants by analyzing the high-frequency data from Oct. 2007 to Oct. 2008. The sample data were dated between Oct. 2007 and Oct. 2008. These warrants’ underlying assets are the same as stocks’. Through this thesis, the existence of the cointegration between both markets will be examined to decide rather VAR or VECM model should be applied. In the end we will use Impulse Response Function and Forecast Error Variance Decomposition to investigate how these two markets influence each other. The conclusion reveals that spots donate warrants and neither of both has explaining ability over the other. This might be related to the immaturity of warrant market in Taiwan. Although the securities with lower transaction cost have stronger price discovery, the volume is also one of the important factors.
