1. INTRODUCTION
1.1 RESEARCH PURPOSE AND MOTIVATION
1.1 RESEARCH PURPOSE AND MOTIVATION
This study discusses the relation between investor sentiment and financial news in Taiwan. Will an investor sentiment and behavior be affected by public information is widely been discussed in prior researches. Some scholars say that the mass media have a significant impact on financial markets since news reports can contribute to the formation of investor expectations. (Carretta et al., 2010) Chen et al (2011) suggests that investor’s reaction could be affected by news coverage; points out the importance of the role of press media in stock markets. Even the messages posted on the Internet;
there is financially relevant information present and those stock messages help predict market volatility. (Antweiler and Frank, 2004) More and more researches have proved the relations between public information and investor sentiment indicating that this research issue still holds a great importance.
Though this issue has been widely discussed, many researches adopt trading volume or other parameters of market activity to measure investor behavior or sentiment, such as Mitchell and Mulherin (1994), Nofsinger (2001) and Tetlock (2007). This study views from an instinct perspective to see the relationship between public information and investor sentiment.
The indicator used to measure investor sentiment by many studies is the Volatility Index (VIX), which originally used to forecast implied volatility of financial market.
Because there is the widespread consensus that the VIX is a barometer of the overall market sentiment as to what concerns investors’ risk appetite, but also on the fact that there are many trading strategies that rely on the VIX index for hedging and speculative purposes. (Fernandes et al., 2014) Seemingly VIX provides an obvious meter of market expectation, and represent investor sentiment. Due to these reasons, this study employs this index as a proxy of investor sentiment as well.
‧
The Chicago Board Option Exchange (CBOE) introduced Volatility Index (VIX) in 1993 to provide a measurement of market implied volatility. The index is constructed by trading price of options of Standard and Poor’s (S&P) and through a computing formula offering investors information about stock market. Since 2003, the underlying index of VIX is changed from the CBOE S&P 100 Index (OEX) to the investor’s state of mind. The CBOE Market Volatility Index offers the market place and academic researchers a new measure of stock market volatility.
There are many studies concerning VIX. Fleming et al. (1995) find a strong negative correlation between VIX change and S&P 100 index returns. That means when VIX climbs up, it not only implies rising market volatility but also a negative effect on stock market performance. Moreover, negative stock market moves are accompanied by larger absolute changes in expected volatility than are positive market moves. (Fleming et al., 1995) When stock index declines continuously, investor loses the confident holding a more pessimistic expectation on market performance. Once VIX gets into an extremely high level, investors would fall into a turmoil state, buying put options no matter the costs. On the contrary, when VIX keeps a low position, it means investor holds a tranquil mood, expecting a smooth volatility on market. Due to the interaction between VIX and stock market, the VIX
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
3
are useful contrarian indicators are contrarian indicators, consistent with the view that periods of extreme fear in the stock market have provided excellent buying opportunities. (Simon and Wiggins, 2001)
Over its fourteen-year history VIX has acted reliably as a fear gauge. High levels of VIX are coincident with high degrees of market turmoil, whether the turmoil is attributable to stock market decline, the threat of war, unexpected change in interest rates, or any number of other newsworthy events. The higher the VIX, the greater the fear. (Whaley, 2000) This is why VIX is also called a fear index, and serves a good measurement for observing investor sentiment on market.
In Taiwan, The Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX), which provided by Taiwan Stock exchange (TWSE), serves as an important index for Taiwan stock market for around 50 years allowing investors to get information about overall market movement and performances conveniently. But the index used to measure Taiwan stock market volatility was introduced later in Dec.
2006. Because the option trading of TAIEX was available by Taiwan Futures Exchange (TAIFEX) after 2001, the startup market was not prosperous. There was not much trading volume and not enough daily information provided by the market for establishing the index. The trading price also wasn’t reliable. At that time, some researches try to construct an index to measure Taiwan stock market volatility on the purpose similar to CBOE VIX on their own developed methodology.
After years of development of option market in Taiwan, in 2006, TAIFEX finally introduced the TAIEX Options Volatility Index (TWVIX). It is compiled according to the new CBOE VIX methodology, using the same formula developed by CBOE. The use has been authorized by S&P. Because TWVIX is aiming at the trading activity of options over TAIEX, investors in Taiwan now can obtain this valuable information to
‧
characteristic similar to VIX, induced by CBOE due to the same methodology.As consequence, this study uses TWVIX as a measurement of investor sentiment in Taiwan due to the evidence of prior studies that VIX could be a representative of investor sentiment. The following figure shows the line graph of TWVIX from Dec.
2006 to Dec. 2014.
Figure 1-1 Line Graph of TWVIX
From the figure, we can briefly know the movements of TWVIX. The extremely high level of fear that the index goes beyond 40 is at the period around 2008 during the global financial crisis. This shows that TWVIX effectively reflects the degree of investor fear.
Being a participant of financial market, investor tries to make a decision of buying or selling assets upon the information available on the market. This means information has a significant effect on investor’s thinking and strategy. The sources of obtaining information could be many channels. Since the information plays an
‧
message. Some most common sources to get information are, for example, financial statements and announcements of a company, expert analytic reports, governmental mandates, macroeconomic announcements, media news by press, and even rumors.
So in order to examining the influence power of information on investors, this study investigates the financial news published by Taiwan news media.
This study adopts TWVIX to measure Taiwan investor sentiment in this research, which is mentioned above. When considering public information, the study chooses news articles as the test object. Because press is widely used as a source for obtain information about financial market, and it is the most approachable way for investors to know what happened on market. The main function of news media is to discover facts, collect information, organize findings, and communicates messages with appropriate tools to people on society. And the mostly being concerned is whether news is reported timely or not. These efforts can only be done by a professional organized system. As a consequence, press media plays a unique role in society.
For information acceptances, people nowadays rely so much on news information.
They will be effect on both mentality and behavior after receiving messages by watching news or reading newspapers. On financial markets, messages communicator still is news media, but the tool to send out information is by financial news. People who care the financial-related events often take part in the financial market as investors. In other words, investor may decide whether to buy or sell stocks by news reports they read.
On financial markets, news media may have the ability of affecting investor’s thinking and behavior. Mitchell and Mulherin (1994) find that the number of Dow Jones announcements and market activity are directly related. There is a direct, robust relation between Dow Jones news stories and stock market activity. Chen et al. (2011) claims higher news coverage decreases the information content of earnings and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
6
reduces market responses to unexpected earnings. The study highlights the importance of financial news in conveying value-related information to the markets.
(Chen et al., 2011) Another proof of media’s influence over financial market is that there is a significant return premium on stocks with no media coverage: One practical implication of the results is that coverage by mass media can play a role in alleviating information problems. (Fang and Peress, 2009) This has the further implication that companies’ media-relations activities can affect their cost of capital. This conclusion implies a critical indication that media can not only sending out messages but also achieve some effect on investors purposely. This may be why there are many studies interesting in topics of media-effect.
By the advanced technology, the number of information people generates and absorbs is growing at explosive speed. Facing the massive quantity of media news, people get more and rely more on it. With the development of financial market and economy, there are more and more trading activities on financial markets. A rational investor often reads the information and adjusts his strategy before he trades. So after he gets the information, how much effect has been made on him to influence his sentiment, as well as his trading behavior would be an importance issue for this study to discuss. With the abundant quantity of these market activities and information changes, people could not ignore the relation between these two elements.
Take a step further of examining media news. In addition to news coverage, news content could also affect the investor to make a different decision after receiving information. Investors react more strongly to negative articles. We felt that the subjectivity of the articles may have had an impact on trading behavior. (Schumaker et al., 2012) So this study also examines the article content of media news, to see words been used could have any different effect on readers. Can positive news report
‧
really comfort people? Does negative news fans up people’s flustered mentality leaving a fear atmosphere on financial markets?
To find any evidence of these issues, this study needs to analyze words and opinions of news article contents. In order to extract information from textual articles, this study has to do further analysis to transfer textual data into an analyzable form.
The purpose of this approach is to capture the feature of news article accurately. This study refers to the methods of Carretta et al. (2011), Davis et al. (2012), Garcia (2013) and Huang et al., (2014). Their studies analyze the textual articles by counting the numbers of words or sentence in certain categories. So the number aggregated or by other calculating process would help the data reflect the nature of those articles. For example, Carretta et al. (2011) defines the number computed by the formula:
𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑤𝑜𝑟𝑑𝑠 − 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑤𝑜𝑟𝑑𝑠
𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑤𝑜𝑟𝑑𝑠 + 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑤𝑜𝑟𝑑𝑠 as a tone measurement. The value obtained is between
−1 (completely negative news) and 1 (completely positive news). The scale (from −1 to 1) allows them to express the degree with which news is positive or negative and the strength of the tone of communication of news. (Carretta et al., 2011)
This study takes a similar method to analyze the tone of news content. But before getting ready to compute the number of words in news content, the study should define what kind of words this research caring about and what categories of words to add up the number with. As for the step before analyzing the content, the study should build up a dictionary containing the categories of words and word lists. The dictionary approach is referring to Carretta et al. (2011), Davis et al. (2012) and Garcia (2013).
Their studies adopt a dictionary which storing words related to their research topic.
That enable their study to process the raw data of textual content into an output of value. So to follow their approach, this study also builds up a dictionary in Mandarin Chinese with Chinese terms. For the purpose of investigating financial news and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
8
investor sentiment, the terms this study brings into the dictionary are used to illustrate people’s mental situation, sentiment and market performance. And for the simplicity, the words sort into only two categories of positive words and negative words, because this study focus on the effect of positive and negative words being used in news content. By then, it allows this study to capture the degree of sentiment expressed in news by counting the good and bad words used by news author. By combining these approaches, this study examines the tone from textual contents of financial news.
Many studies try to explain the media-effect. The power of public information is a critical issue on financial market in modern society. The most common research design is by proving the relation between media activity and financial market behavior.
Ederinton and Lee (1993) find that scheduled macroeconomics news announcements actually do effect on interest rate and foreign exchange future markets. Such is the study talks about the relation between media and market. And those studies mentioned investor sentiment, e.g. Simon and Wiggins (2001), use VIX as an indicator to represent investor sentiment to find relation or possibility of prediction of financial market. This study connects these issues explaining the effect of media news on investor sentiment measures by TWVIX, and combining the text analysis approach to discover if word matters as well.
Since VIX has acted reliably as a fear gauge, high levels of VIX are coincident with high degrees of market turmoil. (Whaley, 2000) Empirical evidence of relationship between VIX and financial news could be useful for describing the strength of message spreading by news media, and how much impact has been made on investors by language spoken by financial news. Comparing to prior researches, we believe this study contributes evidence with a different perspective to explain the media effect on investors in Taiwan.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
Consequently, this study analyzes the financial news articles released by China Times and Commercial Times in Taiwan during Dec. 2006 to Dec. 2014. Examine the importance of media news, word usage and tone of news article by media, to see whether there is the effect on investor sentiment in Taiwan.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10