本文在分盤競價(Call Auction)交易機制的根基上,以前一交易盤成交價為基礎,推論當盤的淨委託量(Net Order),再依據淨委託量定義當盤的委託單失衡(Order Imbalance)方向及程度,進而推導出委託單失衡的衡量方法。應用本推得的委託單失衡衡量方法,所需的資料包括:成交價、揭示買債、揭示賣價、揭示買價下的委託量、揭示賣價下的委託量,這些資料都是即時公開可得的。因此,在沒有臺灣證交所獨有的委託資料情況下,研究人員與一般投資人也能根據本文的方法,反峙的衡量臺灣股市各交易盤的委託單失衡狀態。為了驗證本文委託單失衡衡量方法在臺灣股市的適用性,我們分別以民國93年7月1日到9月30日期間所有上市公司的629萬筆日內交易資料,以及民國85年9月1日至88年4月30日期間市值最大的30家公司之476萬筆日內交易資料進行實證。結果發現,本文的方法能有效的衡量台灣股市之委託單失衡,]衡量所得的當盤委託單失衡狀態可以正確預測下一盤成交償的變動方向。
This study focuses on the order imbalance of call auction market. We base on the net order reached between two auctions to define the order imbalnce (that is the relative size of the buy and sell orders). Based on our definition, we develop the general rules to infer the order imbalance for call auction market. The data needed for applying the rules are: trade price, both the best bid and ask and their sizes. In general, these data are available for all market participants, so investors can measure the order imbalance in time by our rules without the original order data. To verify the effective of our rules, we emply more than ten million call auctions data from Taiwan stock market. The empirical result shows that our rules can effectively measure the order imbalance, and the order imbalance of current call can correctly predict the movement direction of transaction price of next call.