以選擇權理論估計訊息交易機率

淡江大學機構典藏 > 商管學院 > 財務金融學系暨研究所 > 學位論文 > Item 987654321/31505

請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/31505

題名:	以選擇權理論估計訊息交易機率
其他題名:	Measuring the probability of informed trading by option approach
作者:	江孟穎;Chiang, Meng-ying
貢獻者:	淡江大學財務金融學系碩士班段昌文;Duan, Chang-wen
關鍵詞:	逆選擇成本;訊息交易機率;Panel Data模型;選擇權;Adverse selection cost;probability of informed trading;panel data model;Options approach
日期:	2008
上傳時間:	2010-01-11 00:50:34 (UTC+8)
摘要:	本文選取S&P100指數成份股及ETF，觀察2001年1月29日NYSE將報價檔次從分數改為小數報價前後訊息交易機率，並以NASDAQ股票作為對照組，並將成交及報價檔合併使每筆成交皆對應一筆報價，依據 EKOP及選擇權模型估計訊息交易機率，並對日內訊息交易機率的型態作一探討。最後，我們在控制共同變數下，應用Panel Data模型來分析控制變數、訊息代理變數與交易機率間的連結關係。實證顯示，NYSE將報價檔次改為小數報價後，不論是運用EKOP或選擇權模型，均發現市值大、中及小公司所估計出之訊息交易機率是下降的，顯示縮小報價檔次後對於造市者之逆選擇成本是有所改善的，而ETF之訊息交易機率則是上升的；另外，各類公司於小數報價前後之訊息交易機率均在每日剛開盤時為最高，顯示訊息交易者傾向於開盤時即利用其私有資訊來進場交易。以Panel Data模型進行實證的結果顯示，小數報價前大公司與小公司之最小量比率以及中公司之中額比率對訊息交易機率的影響皆為正；最後，透過加入虛擬變數發現在縮小報價檔次後確實會改變訊息代理變數與訊息交易機率的關係，除小公司之中額比率外，各類公司均一致顯示最大量比率、最小量比率與中額比率對訊息交易機率的影響程度增加。 The paper attempts to observe the probability of informed trading around the pre-/post- decimal pricing in January 29, 2001 for S&P100 constituent stocks and ETF by Easley, Kiefer, O’Hara, and Paperman (1996) and option approaches, In cross and time sectional analysis, we examine the linkage between from controlling variables, proxy variables of informed trader and the probability of informed trading through the Panel Data model. Empirical results find that the probability of informed trading is decreasing after decimal pricing on equity sample. It implies that the probability of adverse selection is decreasing on market makers but fails for ETF sample. We also show that informed traders prefer the time between after open of the trading day and midday to the other one about submission order. The empirical results of the Panel Data model illustrate that the relationship between the informed agency variables of trading volume following small-quantitity of trading day and probabilities of informed trading are positive during pre-decimal pricing to high-/low- size stocks, moreover, the trading volume following medium-quantity of trading day is positive with the probability of informed trading for middle-size stocks. Finally, the Panel Data model adding dummy variables, we find that the relationship between the informed agency variables and the probability of informed trading would be change after decimalization of NYSE. Furthermore, large, small, and medium quantities have more effect on the probability of informed trading for equity samples except medium quantities on low- market capitalization companies.
顯示於類別:	[財務金融學系暨研究所] 學位論文

文件中的檔案:

檔案	大小	格式	瀏覽次數
	0Kb	Unknown	342	檢視/開啟

在機構典藏中所有的資料項目都受到原著作權保護.

TAIR相關文章

資料載入中.....