LPPL交易策略應用-探討亞洲主要股市指數

Applications of LPPL(Log-Periodic Power Law) trading strategies on major Asian stock market indices

黃健寧(Huang Jian-Ning)

指導教授 ： 岳夢蘭

國立政治大學/商學院/財務管理學系/碩士(2021年)

https://doi.org/10.6814/NCCU202101135

摘要

本篇論文的模型主要建構於LPPL(Log-Periodic Power Law) (Johansen et al., 2000)。先前的研究者已將LPPL交易策略在全球股市上進行試驗，然而在亞洲市場，尤其是台灣、日本、中國股市上表現不突出或不一致，並認為主因-出在於亞洲股市主要屬於短循環週期(Pham Huu, 2018; Mamageishvili, 2019)。因此本篇論文將調整Mamageishvili的方法，以找出適合亞洲股市的LPPL初始時間窗口(initial time window)。首先利用LPPL在兩種過濾條件:早期預警(Early Warning), 結束標誌(End Flag)下來偵測各自的LPPL泡沫信心指標( LPPL Bubble Confidence Indicator)數值，再取100日移動平均，進而轉換成不同等級的交易訊號，以此作為交易。本研究結果發現台灣、日本、南韓股市指數避開2008年金融海嘯，並在2009年金融海嘯結束後低點準確進場，使得最終結果的各項報酬指標上均優於市場；另一方面，香港、中國上證、中國深圳指數除了準確避開2008年金融海嘯和進場於2009金融海嘯落底之時，並在2018年中美貿易戰中股市重挫時進場，而兩大中國指數更是再次避開2015年中國股災。本篇應證亞洲主要股市指數屬於短循環週期。在亞洲主要股市上，以LPPL短期初始時間窗口(initial time window)方式建構交易策略，可以在大型正(負)向泡沫發生時，抓對時機進出場，勝過買進持有策略並產生alpha。

關鍵字

LPPL交易策略 ； 泡沫 ； 亞洲股市 ； 短週期 ； 初始時間窗口

並列摘要

The model of this thesis is mainly constructed from LPPL(Log-periodic Power Law) (Johansen et al., 2000). Previous researchers have tested LPPL trading strategies on global stock markets, but the performance of Asian markets, especially Taiwan, Japan , and China, is not remarkable or consistent, and the main reason is that Asian stock markets are mainly short cycle (Pham Huu, 2018; Mamageishvili, 2019). Therefore, this thesis will adjust Mamageishvili's method to find the LPPL initial time window suitable for Asian stock markets. First, LPPL is used to detect LPPL Bubble Confidence Indicator values under two filter conditions: Early Warning and End Flag, and then adjust with a 100-day moving average. This translates into different levels of trading signals, which are used as real trade. The results of this study show that Taiwan, Japan, and South Korea stock market indexes avoid the financial crisis of 2008 and enter the bottom of the market after the ending of the financial crisis in 2009 accurately, which makes the final results of all the return and risk/reward ratios are better than the market’s corresponding ones; On the other hand, in addition, to accurately avoiding the financial crisis of 2008 and entering the bottom of the market after the ending of the financial crisis in 2009, Hong Kong, Shanghai, and Shenzhen indexes also entered the market when the stock market plunged in the China–United States trade war in 2018, while the two major Chinese indexes also avoided the Chinese stock market turbulence in 2015. This article should confirm that major Asian stock indexes belong to a short cycle. In major Asian stock markets, a trading strategy constructed with a short-term LPPL initial time window can outperform a buy-and-hold strategy and produce alpha in the event of a large positive (negative) bubble.

並列關鍵字

LPPL ； Bubble ； Asian markets ； Short periodic ； Initial time window

參考文獻

Filimonov, V., and Sornette, D. (2013). “A Stable and Robust Calibration Scheme of the Log-Periodic Power Law Model.” Physica A: Statistical Mechanics and its Applications, Vol. 392, Issues. 17, pp. 3698–3707.

Forró, Z., Woodard, R., and Sornette, D. (2015). “Using trading strategies to detect phase transitions in financial markets.” Physical Review E, Vol. 91, Issues. 4 . 042803

Jacobsson E. (2009). “How to predict crashes in financial markets with the log-periodic power Law” (Master’s Thesis, Department of Mathematical Statistics,

Stockholm University, Sweden). Retrieved from https://www2.math.su.se/matstat/reports/serieb/2009/rep7/report.pdf

Jiang, Z.-Q., Zou, W.-X., Sornette, D., Woodard, R., Bastiaensen, K., and Cauwels, P. (2010). “Bubble diagnosis and prediction of the 2005–2007 and 2008–2009 chinese stock market bubbles.” Journal of Economic Behavior & Organization, Vol. 74, Issues. 3, pp. 149–162.