動態成交量加權平均價格交易策略績效實證-以台積電為例

Abstract

證券市場在80年代後期全面電子化成交以及資訊技術的進步使得程式交易迅速發展。程式交易能使券商或經紀商在資訊流通快速的市場中以自動且迅速的下單策略取得想要的交易部位且大幅降低市場衝擊，進而降低交易成本。VWAP (Volume Weighted Average Price)演算法交易策略即是目前市場中的主流策略之一。本研究以台灣權值股龍頭台積電(2330)和金融股元大金(2885)的逐筆交易資料作為研究對象，使用本文所提出能依據目前的價量資訊作出動態調整的交易策略，以實證動態策略是否擁有比傳統策略更好的績效。更進一步的使用歷史交易數據作統計分析，實證動態策略是否比傳統策略更能降低市場衝擊。

Because of the rapid development in algorithmic trading, brokers can use strategies to get the positions they want automatically and fast. It also reduces the trading cost and its impact to the market where information flows quickly. VWAP (Volume Weighted Average Price) algorithmic trading is one of the main strategies in the stock market. Based on tick trade prices of TWSE (2330) and Yuanta (2885), we developed the strategy which can dynamically modify the VWAP and examined whether this dynamic strategy has better performance than the traditional strategy. Moreover, we apply the historical data to test whether the dynamic strategy can reduce more market impact than the traditional strategy.