Using High Frequency Transaction Data to Estimate Time-Varying Adjusted Probability of Informed Trading.

• Main Authors: Chia-Chun Chen, 陳嘉純
• Other Authors: Shew-Huei Kuo
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/52036813645813625186

碩士 === 國立雲林科技大學 === 財務金融系 === 102 === Probability of informed trading (PIN) presents the degree of asymmetric information. There have a positive relation between asymmetric information and return, but asymmetric information also can result in the loss on uninformed traders. Therefore, PIN is very important for uninformed trader and informed trader. In this paper we use PIN-AACD (Tay et al., 2009) model to estimates the PIN of Taiwan's traditional industries and the electronics industry, a total of 18 listed companies. The empirical studies in Taiwan always use EHO (Easley et al., 2002) model to estimate PIN. But EHO model only considers the aggregate buy and sell order. However PIN-AACD model considers more comprehensively. PIN-AACD model take volumes, trade direction, continuous buy(sell) order into the model. In addition, PIN-AACD model includes AACD (Bauwens and Giot,2000) model when compute PIN, using the stock which have ACD phenomenon to estimate the arrival of buy and sell orders. In the empirical results, we found the standard deviation of PIN of large stocks are less than small stocks. That represents investors will face large risk when they invest in small stocks. Furthermore, we also found OUCC, U-MING, Les enphants, DELTA has higher degree of asymmetric information.