平穩次序訊息比率模型之價格發現研究

Price Discovery with Ordering Robust Information share

陳佳國(Chia-Kuo Chen)

指導教授 ： 辛敬文

共同指導教授 ： 葉錦徽

元智大學/管理學院/財務金融學系/碩士(2008年)

https://doi.org/10.6838/YZU.2008.00089

摘要

價格發現過程指相似金融產品在各市場間流通時，所產生的實際價格情形，此價格將反映各個市場的訊息及效率性。當某市場均衡價格揭露速度快過其他市場時，此市場將具資訊優勢，其投資者可依此來判斷個別市場價格狀況來做自己最有利的投資。現在判別價格發現方法最主要是 Hasbrouck (1995) 以及 Gonzalo, Granger (1995)，但現有的方法都有其缺失。Hasbrouck (1995) 的例子是各市場在短期彼此有關係下，無法算出統一且不受變數次序影響的訊息比率，Gonzalo, Granger (1995) 的算法僅考慮恆常因子均衡的調整，來做為價格發現之依據，但其忽視短期變動的影響。一般來說，在價格發現方法上，Hasbrouck (1995) 的方法比起 Gonzalo, Granger (1995) 更具說服力，因為相較於 Gonzalo, Granger 的方法，Hasbrouck 的方法包含了更多的資訊來為價格發現做解釋，但唯一遺憾的是運用此法所得的訊息比率將因變數順序改變而改變其原本比率，本文將針對這缺失來做改進。本文利用 Koop, Pesaran, Potter (1996) 和 Pesaran, Shin (1998) 一般化衝擊反應函數概念來設計計算訊息比率新方法,在捨棄 Cholesky 分解所導出一般化衝擊反應能成立下，我們應可得到不受變數次序影響的訊息比率。再來,本文也討論 Lien, Shrestha (2008) 所設計修正訊息比率,最後，本文將利用例子和實證來印証其可行性及在價格發現過程所能提昇的相對好處。

關鍵字

價格發現 ； 訊息比率

並列摘要

Price discovery has long been an active research area in empirical finance since it can identify the leading sources of information disclosure among different markets that trade the same or similar financial assets and thus helps in investments. The information share proposed by Hasbrouck (1995) and common factor approach proposed by Gonzalo and Granger (1995) are undoubtedly the most popular ones among many others for price discovery, along with their own merits and demerits. Hasbrouck's approach is persuasive than Gonzalo and Granger's approach since it allows for short-term dynamic information in assessing price discovery. However, once the innovations among markets are correlated, the obtained information shares are not invariant to the ordering of the variables. Though Gonzalo and Granger's approach does not suffer from such ordering problems, their approach cares only information from permanent component and neglects completely the valuable information on market interactions from transitory components. This paper uses the perspective of generalized impulse response function from Koop, Pesaran, and Potter (1996) and Pesaran and Shin (1998) to devise a new method for calculating information share. The new approach goes without using Cholesky decomposition and thus does not suffer from the ordering problem inherited in Hasbrouck (1995). Besides, this paper also discusses the modified information share devised by Lien and Shrestha (2008). Finally, we will use several examples and an empirical application to verify the feasibility and comparative benefits of using the new approach over the other methods in price discovery.

並列關鍵字

price discovery ； information share

參考文獻

11. Lien, D., and Shrestha, K., 2008, A new information share measure and its application to futures and spot markets, Forthcoming in Journal of Futures Markets.

1. Baillie, R.T., Booth, G.G., Tse, Y., and Tatyana, Z., 2002, Price discovery and common factor model, Journal of Financial Markets 5, 309-321.

2. De Jong, F., 2002, Measures of contributions to price discovery: a comparison, Journal of Financial Markets 5, 323-327.

3. Fama, E.F., 1970, Efficient capital markets: a review of theory and empirical work, Journal of Finance 25, 383-417.

4. Gonzalo, J., and Granger, C.W.J., 1995, Estimation of common longmemory components in cointegrated systems, Journal of Business and Economic 13 (1), 27-36.