Print Email Download

Paid Writing Services

More Free Content

Get Your Own Essay

Order Now

Instant Price

Search for an Essay


Semi-strong market efficiency

Working Title:

A systematic approach to assess the semi-strong market efficiency – an analysis of long term relationship between public available information and stock information in Taiwan

Introduction

Public investors usually make their investment decisions according to some or all information that is available to them, such sources including newspaper, TV news, stock brokers, financial websites or even rumour. Some experts encouraged investors to pay more attention to understand their investment portfolio by studying relevant information (e.g. industrial news and financial reports). They believed that by studying past information, investors could improve the process of decision making or even predict the future trends of stocks. In the same time, a classical financial theory also told us that in a market with semi-strong or strong form efficiency, the prices have already reflected all publicly available information. That is to say, investors might be unable to make a profit by analysing all past information they could access. However, some cases always could be found to challenge this theory. This situation confuses many investors when they found the imperfection of financial markets. This proposed research is trying to establish a model to access whether a relationship exists between publicly available information and stock information for any given financial markets and give investors a clear result by using a proven quantitative analysis.

Literature review

The question of whether financial markets can be predicted has been studied by many researchers. Some believed that the behaviour of financial markets are random and unpredictable while some believed that by using information, it is possible to beat the markets and make excessive profits. One famous theory, efficient market hypothesis, is usually mentioned when people discussed this topic. Efficient market hypothesis asserts that while investors purchase stocks, they will use any available information efficiently. All known factors which affect the price of stock have been reflected in the price of the stock (Dimson and Mussavian, 1998). According to EMH theory, investors cannot outperform the market consistently by just using the past, public information.

In 1970, Fama (1970) proposed three types of efficient market form in his research: weak form efficiency, semi-strong form efficiency and strong form efficiency. These three forms are different in the extent of how the market reflects the information. In a market with weak form efficiency, the current stock price has reflected all information provided by past stock price. So, investors cannot predict the future stock price by analysing the past stock price. Based on the random work hypothesis (Malkiel Burton, 1973), it has implied that the future information (future price) will come to us randomly so that investors have no chance to gain excessive profits through past price information (Fama, 1965). In a market with semi-strong weak form efficiency, the current stock price has reflected all publicly available information. Investors relying on the financial reports, economic situations, political circumstances, etc. cannot predict the future stock price or gain excessive profits (Chance, 1985). In a market with strong form efficiency, the current stock price has reflected all publicly available information and all non-publicly information (insider knowledge). Although non-publicly information doesn’t release to public, some investors can still obtain these information by their own channels or connections. In this situation, investors also cannot make excessive profits through inside trading (Rozeff and Zaman, 1988; Finnerty, 1976). However, there has been a dispute over the efficient market hypothesis between researchers, because many evidences reflect the imperfection in financial markets and their inefficiency. Some “behavioural economists” asserted that cognitive biases factors such as confidence, reaction, and information bias result in the inefficient financial markets. This also implied that investors are not always rational (Stout, 2002; Fama, 1998; Rosenberg et al., 1985; Givoly and Lakonishok, 1979; Jensen et al., 1978).

Although researchers realized the fact that there are many factors affecting the efficiency of financial markets, it was still less helpful for investors to make decisions. Investors would like to know whether the markets in which they are interesting are efficient and according to the different characteristics of markets, they can adopt different investment strategies. Some researchers tended to pay attention on financial indices such as dividend and financial ratios and use them to assess the target market. An early study of the efficiency of the UK stock market by Marsh (1979) found that the UK stock market has a semi-strong form efficiency by observing the relationship between the timing of rights issue announcement and share prices(Marsh, 1979). For assessing the efficiency of 15 emerging capital markets, Karemera et al.(1999) evaluated the stochastic properties of local currency and US dollar-based returns. The result showed that most of these emerging markets seemed to be weak form efficient (Karemera et al., 1999). Busse and Clifton Green (2002) described an unique evidence about NYSE’s market efficiency. They noticed that when CNBC TV broadcasted an positive news about a individual stock, the stock price reflected the news within seconds and traders who took action to purchase the stock within 15 seconds have obvious chance to make significant profits (Busse and Clifton Green, 2002). Their evidence showed that to some extent the NYSE fits the semi-strong form efficiency.

Research Question and Aims

This research question of this dissertation will be to provide a systematic approach to explore the relationship between public available information and stocks’ attributes that especially refer to price and volume. As mentioned previously, many literatures proposed their theories and observation to explain the concept of market efficiency. However, most of the evidence which highly depended on few specific financial indicators such as past prices and static financial figures seems to only support that the most of markets are at least weak form efficient. It seems likely that this question cannot be easily answered by just using a single theory and few particular cases to judge to what extent a financial market is semi-strong efficient. Because if one would like to evaluate the semi-strong market efficiency, the problem of handling the numerous publicly information would be the biggest challenge the researchers have to dealt with.

Through using information retrieval technology, the proposed research is aiming to establish a quantitative model that could assess the semi-strong market efficiency of any given financial markets, but due to the time constrain we will only choose Taiwan’s stock exchange as the instance. Firstly, the model should help us to judge whether a correlation exists between the publicly available information and the stock price or volume. Secondly, if a correlation exists, it should give us a more detailed picture about the patterns between them. The most important difference between this proposed research and other studies is that it would not focus on particular figures as others did but analyze all related information on varied media.

Proposed Methodology

It is apparent that the intention of the proposed research is to assess the hypothesis of the relationship. Lancaster (2005) stated that “Deductive research develops theories or hypothesis and then test out these theories or hypothesis through empirical observation.” Using a deductive approach seems reasonable. A research methodology which is more objective and based on facts was the preference for researchers to conduct this proposed research. Owing to the proposed research’s ontological and epistemological orientation (Garner et al., 2009), a quantitative methodology (strategy) with statistical and mathematical techniques should be our choice.

Proposed Method

According to the characteristics of the proposed research, the longitudinal design should be the framework by which researchers conducted it. Using the definition of Bryman and Bell, a longitudinal design is a research:

“on a sample on more than one occasion,……or may involve content analysis of documents relating to different time periods……”(Bryman and Bell, 2007:71)

The proposed research would like to find out whether a relationship exists between publicly available information and stock information, so it needs to collect quantitative data from secondary data through newspapers or other official sources. This longitudinal design is suitable for analysing the variation and the relationship between several different attributes. Under the context of this proposed research, it is appropriate for achieving our aims. One of the advantages of the longitudinal design is that it collects multi-attribute data in a large amount, so it is more likely to be accepted to carry out a research which makes a general conclusion in broad contexts. Combined with quantitative techniques, the advantage would be more obvious.

These following sections are very critical for conducting this research with a quantitative strategy (Singh, 2007).

Data Collecting and Handling

The data information required by this proposed research consists of the public available information and the transaction data of stock markets. The public available information may come from many sources. Firstly, we have to set up the criteria in order to filter out unwanted sources. O'Reilly III (1982) noted that a direct relationship exists between the quality and accessibility of information and the decision making performance. For ensuring the performance of analysis result, we have to be careful what should be included in the criteria. The following criteria may be useful: for quality - authority, coverage, structure, dependency, consistency and relationship; for accessibility – availability and stability.

One of the major information sources is daily news. For covering the majority of investors, we would survey what news sources are investor’s favourites and choose several top newspapers according to the percentage of their population coverage. Another consideration is whether contents of news are stored digitally and could be accessed through Internet. Paper-based information would increase the difficulty of dealing with it.

Other key sources include corporations’ announcement and their financial information (reports). To comply with the requirement of government’s regulation, any public company in Taiwan has to release it financial reports regularly and announce significant information as soon as possible after important events happen.

Transaction data of Taiwan stocks could be gathered from the Taiwan Stock Exchange Corp.’s official site (TWSE, 2010) for free. However, the TWSE does not provide real-time connection to transaction data. If we would like to analyse the data on the fly, some securities companies, like Polaris (Polaris, 2010), could provide the real-time connection service. Technically, in early stage, the stock transaction information also could be transmitted into spreadsheet software for initial analysis.

Expect for stock transaction information, other information usually exists in an unstructured or semi-structured form. Paring and encoding these unstructured information are critical challenges for this stage (Ferrucci and Lally, 2004). We will solve this problem by utilising information technology.

All above information would be downloaded and stored into the dedicated database for further use.

Data Analysis

In order to explore the relationship between the publicly available information and stock transaction information, at least two categories of quantitative analyses would be used(Wisniewski, 2006; Bierman et al., 1991). The first category is statistics. The useful statistical techniques for the proposed research may include correlation coefficient for correlation, time series for trends, logistic regression for model, and factor analysis for focusing on critical attributes (Anderson et al., 2008). The detailed parameters and calculation need to be decided after. The second category is data mining. Considering the requirement of the proposed research, classification/decision tree mining is suitable to train a model and to predict the possible future stock outcome (Apte and Weiss, 1997).

Work Plan

April

Collect and analyze candidate available information.

Analyze the data format of available information provided by Taiwan stock exchange.

Survey appropriate algorithms for categorizing unstructured or semi-structured data.

Decide the detailed quantitative models for analysing the relationship.

Discuss with supervisor.

May

Design the data model.

Design the architecture of the information scoring system.

Implement the prototype of information scoring system.

Implement the chosen quantitative models.

Proof of concept.

June

Clean and Input data into the system.

Analyze the data and retune the parameters.

Start to write up the dissertation.

July

Write up the findings, discussion and conclusion.

Discuss with supervisor.

Revise the dissertation.

August

Buffer

Ethical Issues

The ethical issues in the proposed research seem relatively minor. The major reason is that all needed information and data would be gathered from those sources that are available to public. So, it is not necessary for researchers to co-operate with any gatekeepers for access to information sources (Bruckman, 2002). However, if this research progress successfully, the findings of this research may challenge some investor’s perception of financial markets. Whether it would induce psychological stress or anxiety is still uncertain at the moment (Smock, 1955).

Another ethical issue we could consider is the standard of scoring information. Although the way how to judge the information seems not to be harmful to any individual, it would still pass judgment on the performance of corporate entities, which are comprised of people . Some corporations may not be willing to be judged publicly even though it is totally legal. The only thing we could do is to ensure to score the information in an impartial manner as much as possible (Carse, 1998).

So far the proposed research focuses on the assessment of semi-strong form efficiency so that the only needed information are freely obtained from public sources. Nonetheless, in the long term, if we consider evaluating the strong form efficiency of markets, the insider knowledge would be one of the biggest ethical issues. The leakages of insider knowledge are strictly prohibited among most of countries around the world (Beams et al., 2003). The difficulty of gathering insider information may also explain the reason why few studies have been conducted in this area.

Limitations and Contingency Plans

First of all, there are a large number of types of information sources available to investors, but it is unlikely to collect and analyze all these information in the period of conducting the research. Under the situation of limited resources, we would be likely to consider only major mainstream media having large coverage in Taiwan. It may lose correctness to some extent, but maintain its feasibility.

The second limitation would be the interval between publishing information and quoting prices. If the interval of the most likely available data is one day, the phenomena which could be observed may be limited to a daily, larger scale. The information of patterns of relationship within a small scale, one day, might lose. For example, it will be more difficult to analyze the behaviour of day traders, who trade their stocks in a relatively very short period. However, this limitation should not make any obvious difference on the model. Once the real-time sources are available, this limitation will not exist anymore.

The third limitation is that because most of available information exist or are provided in an unstructured or semi-structured from, it would be a challenge to transform the information into the structured data, which could be efficiently dealt with through quantitative methods.

Perhaps, the most important limitation is that the result of analysis would be highly affected by the individual difference of perception of the information. The standard of scoring the information depends on the subjective judgment, and in some cases, it may not be consistent for each company if we score the information manually. Considering the balance between the time of preparing the data and the quality of the data, we might decrease the number of the companies chosen to be analyzed. This contingency plan would not influence the validity and the reliability of the result. The critical mission of this research should be to establish a reliable model rather than to cover all companies as many as possible.

Further and Future

As I mentioned in the limitation section, constrained by limited resource, the research is very likely to adopt a manual approach to score the information. However, it would be helpful for practical usage if some automatic scoring mechanisms can be established. The further improvement should rely on the progress of information theory in cross-areas instead of pure financial area.

The outcome of this research would not only reveal the efficiency of specific markets (e.g. Taiwan stock market) or specific stocks but also develop a system that could be applied to other markets from which information would be imported. In the design phase, the architecture and framework would be designed carefully for its expandability and scalability in order to be possible for other researchers to build their own “information scoring algorithm” or “information indices” and plug into the model the proposed research will build. Moreover, after a strict verification, the model could also be integrated into other statistics systems and financial programming transaction systems as indicators. It may raise the possibility of economic value in the future.

References

Anderson, D., Sweeney, D. and Williams, T. (2008) Statistics for business and economics. South-Western Pub.

Apte, C. and Weiss, S. (1997) 'Data mining with decision trees and decision rules', FGCS. Future generations computer systems, 13, (2-3), pp. 197-210.

Beams, J., Brown, R. and Killough, L. (2003) 'An experiment testing the determinants of non-compliance with insider trading laws', Journal of Business Ethics, 45, (4), pp. 309-323.

Bierman, H., Bonini, C. and Hausman, W. (1991) Quantitative analysis for business decisions. Richard d Irwin.

Bruckman, A. (2002) 'Ethical guidelines for research online', Retrieved June, 9, pp. 2003.

Bryman, A. and Bell, E. (2007) Business research methods. Oxford University Press, USA.

Busse, J. and Clifton Green, T. (2002) 'Market efficiency in real time', Journal of Financial Economics, 65, (3), pp. 415-437.

Carse, A. (1998) 'Impartial principle and moral context: securing a place for the particular in ethical theory', Journal of Medicine & Philosophy, 23, (2), pp. 153.

Chance, D. (1985) 'A semi-strong form test of the efficiency of the treasury bond futures market', Journal of Futures Markets (pre-1986), 5, pp. 3.

Dimson, E. and Mussavian, M. (1998) 'A brief history of market efficiency', European Financial Management, 4, (1), pp. 91-103.

Fama, E. (1965) 'Random walks in stock market prices', Financial Analysts Journal, 21, (5), pp. 55-59.

Fama, E. (1998) 'Market efficiency, long-term returns, and behavioral finance', Journal of Financial Economics, 49, (3), pp. 283-306.

Fama, E. F. (1970) 'Efficient Capital Markets: A Review of Theory and Empirical Work', The Journal of Finance, 25, (2), pp. 383-417.

Ferrucci, D. and Lally, A. (2004) 'UIMA: an architectural approach to unstructured information processing in the corporate research environment', Natural Language Engineering, 10, (3-4), pp. 327-348.

Finnerty, J. (1976) 'Insiders and market efficiency', The Journal of Finance, 31, (4), pp. 1141-1148.

Garner, M., Wagner, C. and Kawulich, B. (2009) Teaching research methods in the social sciences. Farnham, Surrey, England; Burlington, VT: Ashgate Pub.

Givoly, D. and Lakonishok, J. (1979) 'The information content of financial analysts' forecasts of earnings:: Some evidence on semi-strong inefficiency', Journal of Accounting and Economics, 1, (3), pp. 165-185.

Jensen, M., Field, S. and Park, T. (1978) 'Some anomalous evidence regarding market efficiency', Journal of Financial Economics, 6, (2/3), pp. 95-101.

Karemera, D., Ojah, K. and Cole, J. (1999) 'Random walks and market efficiency tests: Evidence from emerging equity markets', Review of Quantitative Finance and Accounting, 13, (2), pp. 171-188.

Lancaster, G. (2005) Research methods in management: a concise introduction to research in management and business consultancy. Butterworth-Heinemann.

Malkiel Burton, G. (1973) 'A Random Walk Down Wall Street', Toronto: George J. McLeod Limited.

Marsh, P. (1979) 'Equity rights issues and the efficiency of the UK stock market', The Journal of Finance, 34, (4), pp. 839-862.

O'Reilly III, C. (1982) 'Variations in decision makers' use of information sources: The impact of quality and accessibility of information', Academy of Management Journal, pp. 756-771.

Polaris (2010) Polaris Financial Group. Available at: http://www.polaris.com.tw/Finairport/default (Accessed: 19 March 2010).

Rosenberg, B., Reid, K. and Lanstein, R. (1985) ' Persuasive evidence of market inefficiency', Journal of Portfolio Management, 11, (3), pp. 9-16.

Rozeff, M. and Zaman, M. (1988) 'Market efficiency and insider trading: New evidence', The Journal of Business, 61, (1), pp. 25-44.

Singh, K. (2007) Quantitative social research methods. Sage.

Smock, C. (1955) 'The influence of stress on the perception of incongruity', Journal of Abnormal and Social Psychology, 50, (3), pp. 354-356.

Stout, L. (2002) 'Mechanisms of Market Inefficiency: An Introduction to the New Finance, The', J. Corp. L., 28, pp. 635.

TWSE (2010) Taiwan Stock Exchange Corp. Available at: http://www.twse.com.tw/en/ (Accessed: 19 March 2010).

Wisniewski, M. (2006) Quantitative methods for decision makers. Prentice Hall.

Print Email Download

Share This Essay

Did you find this essay useful? Share this essay with your friends and you could win £20 worth of Amazon vouchers. One winner chosen at random each month.

Request Removal

If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal:

Request the removal of this essay.


More from UK Essays

Need help with your essay?

We offer a bespoke essay writing service and can produce an essay to your exact requirements, written by one of our expert academic writing team. Simply click on the button below to order your essay, you will see an instant price based on your specific needs before the order is processed:

Order an Essay - via our secure order system!