三種虛擬貨幣之比較與股市關聯性之研究

Comparison of Three Virtual Currencies with Stock Markets

莊秋欣(Chiu-Hsin Chuang)

指導教授 ： 胡為善

中原大學/商學院/企業管理研究所/碩士(2017年)

https://doi.org/10.6840/cycu201700091

摘要

近十年來，金融科技(FinTech)在全球引發流行熱潮，且其六大模式在各產業中均為普遍應用，臺灣也在2015年由全家與BitoEX合作，在「Famiport」上可以購買比特幣，也可將比特幣轉換為消費券，得在全家便利商店使用。日本財務省和金融廳更是將比特幣定位與貨幣相同，日本民眾可以使用比特幣支付電費，且預計在2017年取消購買比特幣等虛擬貨幣所產生的消費稅，以便促進比特幣等虛擬貨幣的發展。
　　本研究探討三個虛擬貨幣間的比較，及其與六個股市之間的關係，其研究標的為：(1)三個虛擬貨幣包括：比特幣、萊特幣及狗幣及(2)六個股價指數包括：S&P500、倫敦金融時報指數、日經225指數、上海A股指數、香港恒生指數及臺灣發行量加權指數。研究期間依比特幣的走勢分為三期，並利用時間序列模型進行虛擬貨幣與股市間之關聯性探討，經由實證結果得到以下結論：
1.本研究發現在第一段期間內，比特幣、萊特幣及狗幣與臺灣發行量加權指數之間並未存在共整合關係；而在第二段期間內，比特幣、萊特幣及狗幣與S&P500、倫敦金融時報指數、日經225指數、上海A股及臺灣發行量加權指數亦無存在共整合關係；但在全樣本期間及第三段期間內，比特幣、萊特幣及狗幣與各股價指數均存在長期穩定之均衡關係；此外，在第二段期間內，比特幣、萊特幣及狗幣與香港恒生指數存在長期穩定之均衡關係；
2.本研究透過因果檢定發現，在全樣本期間、第一段期間及第二段期間內，除虛擬貨幣與S&P500之因果關係模型外，比特幣皆領先於萊特幣，表示比特幣對於萊特幣為領頭者角色，萊特幣會受比特幣波動之影響；而在第三段期間，S&P500指數領先比特幣，其原因可能是美國總統大選造成美國股市震盪，而比特幣被視為避險工具。

關鍵字

Granger因果關係 ； VAR模型 ； VECM模型 ； 比特幣 ； 狗幣 ； 股市 ； 萊特幣 ； 虛擬貨幣

並列摘要

For the past decade, fincancial technology has been very popular in the world, their six models have been applied in various industries. Even in Taiwan, Family Mart cooperated with BitoEX since 2015. Taiwanese now can buy Bitcoin on "Famiport" or convert it to a coupon which can be used at that convenience store. Additionally, Japan's Finance Ministry and the Financial Services Agency treat Bitcoin as the same position with currency. The Japanese can use Bitcoin to pay utility expenses. Japan plans to cancel the consumption taxes on Bitcoin by the end of 2017 in order to accelerate the usage of Bitcoin.
This study examines the relationship between three virtual currencies with the six stock indexes. The targets are as follows: (1) Three virtual currencies: Bitcoin, Litecoin and Dogecoin; (2) Six Stock Indexes: S&P500 index, London’s Financial Times Ordinary Shares Index, Japan’s Nikkei 225 Index, Shanghai’s SSE Composite Index, Hong Kong’s Hang Seng Index, and Taiwan’s TAIEX. This investigation further divides the entire sample period into three sub-periods, and uses the time series model to measure the correlation between the virtual currencies and the stock indexes. Empirical results are summarized below:
1.This study finds that the long-term equilibrium relationship does not exist for the first sub-period and the second sub-period for these virtual currencies with S&P500 index, Financial Times Ordinary Shares Index, Nikkei 225, SSE Composite Index, and TAIEX. However, this investigation finds that three virtual currencies and six stock indexes have co-integration during the entire sample period and the third sub-period. Additionally, this work finds that these three virtual currencies have co-integration with Hang Seng Index during the second sub-period.
2.This study then uses the Granger causality test to find that, during the entire sample period, the first sub-priod and the second sub-period, Bitcoin affects the Litecoin except for the S&P500 index. S&P500 index also affects Bitcoin during the third sub-period, one possible explanation is that the third sub-period may be is affected by US general election.

並列關鍵字

Bitcoin ； Dogecoin ； Granger causality test ； Litecoin ； VAR ； VECM ； Virtual Currency and Stock Indexes

參考文獻

楊馥安，2015，「虛擬貨幣的可投資性研究-以比特幣為例」，中原大學企業管理研究所碩士論文。

李芸綺，2014，「虛擬貨幣發展之探討」，台灣經濟研究月刊37卷1期，頁89-104。

The Economist Special Report. (2015). "Slings and Arrows." The Economist. May, 2015.

Engle, R. F., & Granger, C. W. J. (1987). "Co-integration and Error Correction: Representation, Estimation, and Testing." Econometrica, 55(2), 251-276.

European Central Bank. (2012). "Virtual Currency Schemes. " ECB.