1.1 Background and Motivation of This Research
Investors are looking forward to taking advantage of a price spread between two or more financial markets, such as bonds, stocks, derivatives, commodities and currencies markets. Nowadays, the stock market provides investors an easy-to-liquidate, easy-to-trade market, so investing stocks is one of the most popular investment instruments. Not only the investors, traders but also top tier analysts of investment banking companies are developing many methods to forecast the securities prices, including quantitative and experiential methods. Quantitative methods are based on a mathematical model built to describe the market behaviors. By statistical techniques, a regression curve can be found to fit the historical data with a minimum difference, however, the explanatory power depends upon the adequate selection of independent variables. In statistics, several methods taught us how to select the adequate independent variables, such as forward selection, backward selection, and stepwise selection methods.
Technical analysis of historical data is an example of experiential methods. Ralph Elliot proposed Wave Theory to explain the United States stock markets in 1930s. Elliot
found that the stock prices changed with an observable pattern, and which duplicates in forms of price performance not in price range or time span. No matter which method is employed, the objective is identical, to gain money from the stock markets.
Unfortunately, no one can predict the real price of any stock tomorrow. Because the stock exchange lists the companies owned publicly, the price of the stock is negotiated by the bid and ask price. The bid price stands for the price which buyers are willing to pay, and the ask price is the price at which sellers are willing to sell. The next unpredictable step is called Random Walk in finance. Random walk is a mathematical formalization of a trajectory that consists of taking successive steps in random directions. It implies that the price is like a walking drunkard, the next step depends on the present place he stands in, and we will never know where he will step ahead.
The logarithm of stock price follows a continuous random walk, see Lo and MacKinlay (1988). That is to say, it is impossible to forecast the next step of stock price.
Because of the different time zones, there are always stock markets trading all day long around the world. A phenomenon can be observed that the linkages among the stock markets around the world are getting stronger and stronger, especially the dependence upon the stock market indices of the United States of America. Many academic studies show that the spillover effects of return and volatility exist between
the markets, see Darbar and Deb (1997) and Chou, Lin and Wu (1999).
I have cared about the Taiwan stock markets for several years; I could find that the relationship of securities markets between Taiwan and the United States was getting stronger. With the high transparency of international events and trading activities, the events happened in the United States usually could impact the open price of Taiwan stock market. The open price means the first traded price of the trading day. Therefore, I want to study the correlation of the securities markets between Taiwan and the United States and expect to propose a trading strategy based on the correlation.
1.2 Research Objective
Because of the increasing foreign participation in the Taiwan stock market, the linkages between Taiwan stock market and other countries are strengthened, especially the United States stock markets. Spillover effect1 will be studied in this article, including the price change and volatility spillover effects between Taiwan and the United States markets, and correlation between the two countries’ markets will be modeled as well.
The exchange trading indices include several industries, such as manufacturing, information technology, electronics and etc. For example, the United States Dow
1 Spillover effects are externalities of economic activity or process upon those who are not directly
Jones index covers 30 listing manufacturing companies, and NASDAQ Composite Index covers over 3600 high-tech companies. Each index stands for its specific industry performance respectively. The chosen industry of this research is the electronic industry. The reason why electronic industry selected is that the electronic industry is a representative one in Taiwan and its traded volume of electronic industry sector is weighted by around 60% of the daily total traded volume of TAIEX. So it is reasonable to choose electronic industry sector on behalf of Taiwan stock market performance. The index chosen is NASDAQ Composite Index (abbreviated as NASDAQ) for the United States stock market and Electronic Sector Index Futures (abbreviated as EXF2) for the Taiwan market. As the name said, the underlying index of EXF is the Taiwan Stock Exchange Electronic Sector Index. However, the Taiwan Exchange Index (TAIEX) is employed to compare with NASDAQ, too. The research results will show two pair-wise correlations, NASDAQ with EXF and NASDAQ with TAIEX.
The objectives of this research are:
1. The price changes and volatility spillover effect between Taiwan and the United
2 The trading hours of EXF are from 08:45AM to 1:45PM Taiwan time, Monday through Friday of the regular business days of the Taiwan Stock Exchange. EXF has a daily price limit of +/- 7% of previous day’s close price.
States markets.
2. The correlation of NASDAQ and EXF.
3. The correlation of NASDAQ and TAIEX.
4. Extreme return effects of NASDAQ on EXF and TAIEX.
1.3 Limitation of the Research
The chosen objects, NASDAQ and EXF, are one of the many indices in the United States and Taiwan, respectively. This research provides a method to analyze the correlation between two stock markets, but not a general result for all the markets in the world. Even if for NASDAQ and EXF, the results might be different from the empirical results in this article, because of the time period covered in the sample data.
The properties of the data employed in this article will be shown in chapter 4.
1.4 Framework of the Research
The framework of this research article is structured as the flow chart below.
1. Introduction