An empirical study of efficient market hypothesis and its existence in virtual markets

Contents Page

Declaration and Word Count

Abstract

Acknowledgements

Contents Page

A List of Figures:

A List of Tables:

Chapter 1 - Introduction
1.1 Origin of Efficient Market Hypothesis
1.2 Aims and Objectives of the Research
1.3 Outline of the Chapters

Chapter 2 - Literature Review
2.1 Introduction
2.2 Market Efficiencies
2.3 Information Announcements in Capital Markets
2.4 Anomalies in Capital Markets
2.5 Intra-Day Effect
2.6 Judgments and Values in Computer Games
2.7 Common Exchange Mechanisms
2.8 Regulation of Virtual Markets
2.9 Chapter Summary

Chapter 3 - Methodology
3.1 Introduction
3.2 Research Questions
3.3 Quota Sampling
3.4 Method of Data Collection
3.5 Method of Data Analysis
3.6 Limitations and Validity of the Research Method
3.7 Ethical Consideration
3.8 Chapter Summary

Chapter 4 - Findings and Analysis
4.1 Introduction
4.2 Expected Results
4.3 Actual Results
4.4 Chapter Summary

Chapter 5 - Discussion
5.1 Introduction
5.2 Do in-game virtual markets suffer from the Intra-Day Effect?
5.3 What happens when new information becomes publicly available?
5.4 Does this reflect what happens in real world markets?

Chapter 6 - Conclusion
6.1 Implication of this Research
6.2 Limitations and Further Research

References

Appendix A – Reflective Statement

Appendix B – EMH Timeline

Appendix C – Player ID’s

Appendix D - Formulas

Appendix E – Percentage Changes

Last Page

Declaration and Word Count

I declare the following:-

(1) That the material contained in this dissertation is the result of my own work, that due acknowledgement has been given in the bibliography, and references to ALL sources be they printed, electronic or personal.
(2) The Word Count of this Dissertation is: 9,608
(3) that unless this dissertation has been confirmed as confidential, I agree to an entire electronic copy or sections of the dissertation to being placed on Blackboard, if deemed appropriate, to allow future students the opportunity to see examples of past dissertations. I understand that if displayed on Blackboard it would be made available for no longer, than five years and those students would be able to print off copies or download. The authorship would remain anonymous.
(4) I agree to my dissertation being submitted to a plagiarism detection service, where it will be stored in a database and compared against work submitted from this or any other School or from other institutions using the service. In the event of the service detecting a high degree of similarity between content within the service this will be reported back to my supervisor and second marker, who may decide to undertake further investigation which may ultimately lead to disciplinary actions, should instances of plagiarism be detected.
(5) I have read the University Policy Statement on Ethics in Research and Consultancy and the Policy for Informed Consent in Research and Consultancy and I declare that ethical issues have been considered and taken into account in this research.
(6) I have read the University Policy Statement on Data Protection in Research and Consultancy and I declare that the data collected for use in this dissertation has been properly safeguarded and will be destroyed once the dissertation or subsequent research activity has been concluded. I acknowledge that it is my responsibility to destroy the information with due regard to confidentiality.

SIGNED: [illustration not visible in this excerpt]

DATE: 28.04.15

illustration not visible in this excerpt

Abstract

Virtual and computer games are rapidly increasing with the introduction of the smartphone and the app stores across multiple platforms and devices with an increase in games with virtual economies. This dissertation will analyse the efficient market hypothesis, along with commonly known anomalies and information announcements. It will find out whether there are market inefficiencies in virtual games in the form of anomalies, more specifically the intra-day effect. The intra-day effect anomaly is one of many critiques of the efficient market hypothesis and there have been many studies conducted into the intra-day effect. Most research on the intra-day effect anomaly is concerning real world markets and the results have contradicted one another.

This study looks at the price change movements of 118 randomly quota sampled player cards within the market of FIFA Ultimate Team. Statistical analysis in the form of mean, standard deviation, and coefficients of variances tests were carried out to identify if there were any market anomalies and reactions to information announcements. A strong correlation between market inefficiencies, anomalies, and information announcements had been discovered within the research of the virtual market in FIFA Ultimate Team. The study actually found that because of an information announcement overreaction and an intra-day effect, at a specific time during a Wednesday, a player could sell their card for potentially 233% more than what they could have an hour earlier. This research study in turn supports that market anomalies do exist in games but it was also discovered that the market is semi-strong form efficient in its reaction post-information announcement.

Acknowledgements

First thanks go to my supervisor, Dr David Grundy of Northumbria University for his expertise and advisory on this dissertation. I am grateful for his dedication and contributions towards my dissertation as of his experience with the topic of research. Dr David Grundy was also the finance lecturer for the module from which I developed the initial idea and his lectures significantly helped with understanding and building around the core hypothesis in question.

Another person whom I would like to thank is the Administrator of the website www.futbin.com, who happily provided me with the medium to collect secondary data for the analysis conducted in this dissertation from their system. If it were not for this person, I would not have been able to as easily and as efficiently be able to conduct the analysis within the permitted deadlines of this dissertation.

A List of Figures:

Figure 1: A Venn diagram to show how the three forms of Market Efficiencies are made up.

Figure 2: Standard Deviation of Market Return across Days (by minute, overnight returns included) (Wood et al. , 1985)

Figure 3: Intra-daily Patterns in Stock Returns (Harris, 1986)

Figure 4: The Mean Intra-day 15-minute returns (Bildik, 2001)

Figure 5: Average Sale Price per hour over 6 days (04/02/2015 - 09/02/2015)

Figure 6: Average Sale Price change percentage per hour over 6 day

Figure 7: Standard Deviation of Sale Price per hour over 6 days (04/02/2015 - 09/02/2015)

Figure 8: Standard Deviation of Sale Price change percentage per hour over Wednesday

Figure 9: Standard Deviation of Sale Price change percentage per hour over 5 days

Figure 10: Coefficient Variance of Sale Price per hour over 6 days (04/02/2015 - 09/02/2015)

Figure 11: Coefficient Variance of Sale Price change percentage per hour over Wednesday

Figure 12: Coefficient Variance of Sale Price change percentage per hour over 5 days

A List of Tables:

Table 1: A Table to show what makes up the three forms of Market Efficiencies.

Table 2: A Table to show a list of Information Announcement cases and their main contributors

Table 3: A Table to show a list of anomalies and their main contributors

Table 4: A Table to show the main theorists regarding the Intra-Day Effect

Table 5: Ten Exchange Mechanisms that implement markets

Table 6: A Table to show the sample breakdown of population size and quota per category (Data correct as at 05/01/15).

Table 7: Mean card prices at each hour of each day (04/02/2015 – 09/02/2015)

Table 8: Standard deviation of card prices at each hour of each day (04/02/2015 – 09/02/2015)

Table 9: Coefficient Variances of card prices at each hour of each day (04/02/2015 – 09/02/2015)

Chapter 1 - Introduction

“An ‘efficient’ market is defined as a market where there are large numbers of rational, profit-maximizer’s actively competing, with each trying to predict future market values of individual securities, and where important current information is almost freely available to all participants” (Fama, 1965). The aims of this dissertation are to analyse the efficient market hypothesis theory paying particular attention to market anomalies, information announcements and identifying whether or not it applies within virtual economies. Research will be conducted on a virtual market built within the FIFA Football game franchise. Findings in this dissertation will allow economists to help game designers build efficient virtual economies by potentially highlighting anomalies and abnormal patterns concerning card prices.

1.1 Origin of Efficient Market Hypothesis

Early development of Efficient Market Hypothesis, or EMH first derived from a statement that was published in The Stock Exchanges of London, Paris, and New York in 1889 by George Gibson. Gibson (1889) said, “When shares become publicly known in an open market, the value which they acquire may be regarded as the judgment of the best intelligence concerning them.” (p. 31). This is still the foundation of a contradiction between financial literatures more than 100 years on. The Efficient Market Hypothesis, or EMH, is the theory established on the assumption that prices obey a ‘random walk’ in a speculative market whilst returns are distributed normally. Bachelier (1900) first pioneered this ‘random walk’ in 1900. This was vital to Fama’s (1965; 1969) contribution to Efficient Market Hypothesis.

Frederick MacCauly (1925) criticised a theorist who claimed that his company could predict accurate stock prices. Vance (as cited in Schulz, 1925) said, “We consider the service, in spite of its lack of present perfection, has accomplished far in excess of what is accomplished by the average buyer of securities, and hence its use is justified." (p. 248). MacCauly (as cited in Schulz, 1925) stated that he had “observed a striking similarity between the fluctuations of a stock market and those of a chance curve which may be obtained by throwing a dice” (p. 248). This supports the random walk theory pioneered by Bachelier (1900) some 25 years prior.

The Random Walk was developed and tested further by Fama (1965). At the time, the current methods of predicting stock prices were technical theories and the theory of fundamental value analysis. Both of these theories relied on the assumption that history repeats itself and stock prices will recur in the future (Fama, 1965) and as previously proven wrong by MacCauley (as cited in Schulz, 1925). The random walk cast doubts over these theories as most simply the theory of random walks implies that a series of stock price changes has no memory. The history of the series cannot be used to predict the future in any meaningful way. The future path of the price level of a security is no more predictable than the path of a series of cumulated random numbers (Fama, 1965).

Fama (1970) based his Efficient Market Hypothesis on a subset of information as he proposed three significantly distinguished levels of efficient markets. These were built on assumptions similar to that of Gibson (1889). The three types of markets were weak-form efficient, Semi-Strong Form efficient and Strong Form Efficient (Fama E. , 1970). A full timeline of EMH theorists can be found under Appendix A.

1.2 Aims and Objectives of the Research

The questions derived from the research rationale are; “Do in-game virtual markets suffer from the Intra-Day Effect?” “What happens when new information becomes publicly available?” and “Does this anomaly reflect what happens in real world markets?” In order to answer these questions, research into an online virtual market must be conducted to draw comparisons to real world markets and situations. There are currently minimal, if any, research papers regarding this topic. This research is there for vital for economists who work with game designers to produce and build efficient marketplaces that can be monetised.

1.3 Outline of the Chapters

The dissertation will begin with a review of existing literature regarding efficient market hypothesis paying particular attention to the semi-strong form of efficiency and information announcements. Next, the concept of capital market anomalies, a review of existing anomaly tests regarding real world markets and more importantly the intra-day effect. This is followed by a study of judgements and values in computer games, finished with how in game virtual markets operate, and how they are regulated. The dissertation then goes on to discuss the methods in which analysis will be carried out to generate the results needed to answer the questions set out previously with justification on why these methods are relevant. This is followed by the findings from the research, both expected and actual, which then leads on to the discussion of the findings and whether or not the aims have been achieved. Finally, a concluding chapter summarises the dissertation, and points out objectives for further study on the topic.

Chapter 2 - Literature Review

2.1 Introduction

This section of the dissertation will focus on critically analysing pre-existing research relating to the efficient market hypothesis, market anomalies, and more specifically the Intra Day Effect. The literature review will study tests on real world capital markets carried out by previous theorists in order to pave the way for carrying out a similar test under virtual in-game market conditions. Analysis will also be carried out in the fields of information announcements and the structure of in game markets to have a greater understanding of the outcomes of this research.

2.2 Market Efficiencies

Market Efficiencies were a product of Fama’s (1970) research. He identified three distinct forms based on a combination of certain criteria. These were Weak Form, Semi-Strong Form, and Strong Form Efficiency.

Weak-Form Efficiency (A)

Fama (1970) said that weak-Form Efficiency assumes that all historical stock prices and volume are represented by the current market price of the stock. (p. 309) From Table 1 below and Figure 1 below it shows that A, Historical data, on its own is the single component required for Weak-Form Efficiency shown as number 1.

Semi-Strong Form Efficiency (A + B)

Semi Strong Form Efficiency is the more common form of efficiency because it is how real world markets are perceived but not always, how they are. Semi-Strong Form Efficiency assumes that the current market price of the stock option represents all information available to the public. The only way to beat the semi strong form market would be via insider trading according to Fama (1970 p. 404). This is not always the case as this paper intends to find out.

Table 1 and Figure 1 show that there are two components widely accepted by Fama’s (1965) supporters to achieve Semi-Strong Form Efficiency. These are represented as A, Historical Data and B, Public Information. This creates number 2 on the Table and Venn diagram. Semi-strong-form efficiency implies that neither fundamental analysis nor technical analysis techniques will be able to produce excess returns. There are many critiques to this theory as many leading investment firms and hedge funds reject this notion and have proven to beat the market statistically as seen with Pershing Square's Bill Ackman (Mitchell, 2010). To test for semi-strong-form efficiency, the adjustments to previously unknown news must be of a reasonable size and must be instantaneous. To test for this, consistent upward or downward adjustments after the initial change must be looked for (See 8.4 Information Announcements in Capital Markets). If there are any such adjustments it would suggest that investors had interpreted the information in a biased fashion and hence in an inefficient manner.

Semi-Strong Form Efficiency should be the norm for all world markets in both game and real world. Insider trading is illegal in real world markets yet it does happen as seen with R Foster Winans (The Winans Decision, 1987) and therefor once again contradicts Fama’s (1965) efficient market hypothesis. Insider trading in game markets is considered unethical and usually violates the game developer’s terms and conditions. Unfair advantages are not intentions for any game developer as it hinders profitability for them. (Lehdonvirta & Castronova, 2014)

Strong Form Efficiency (A + B + C)

Strong-Form Efficiency assumes that the current market price of the stock option represents all information both publicly and privately available (Fama, 1970, p. 409). Taking a third look at Table 1 and Figure 1, we can see that number 3, Strong Form Efficiency is made up of components A; Historic Data, B; Public information and C; Private information. Insider trading would not be able to generate excessive returns in this form of market.

illustration not visible in this excerpt

Figure 1 : A Venn diagram to show how the three forms of Market Efficiencies are made up.

illustration not visible in this excerpt

Table 1: A Table to show what makes up the three forms of Market Efficiencies.

2.3 Information Announcements in Capital Markets

The efficient markets hypothesis, more specifically Semi-Strong Form Efficiency mentioned previously suggests that unexpected earnings should be fully incorporated into asset prices soon after being publicly announced. Information announcements comprise of many different forms. Common forms of information announcements are Stock and Dividend related press releases, new product announcements, earnings announcements and operating news announcements (Christensen, Smith, & Stuerke, 2004).

Information announcements can have a direct effect on the share price of the companies highlighted in the announcements. If the market is, semi-strong form efficient as suggested by Fama (1965) the price should adjust to incorporate the new information within its price. However if the market is not efficient, it would be expected to see traders achieve abnormal returns. There has been significant research into the areas of information announcements and a table comprising some key contributors and a summary of their findings are presented below.

illustration not visible in this excerpt

Table 2: A Table to show a list of Information Announcement cases and their main contributors

It appears that investors speculate or hedge prior to the information announcement and the share price struggles to find equilibrium. After the announcements have been made, the price then has to readjust based on how investors view the company because of the effects of the announcement both long term and short term. From the summaries in the table above, a consensus appears to be that there is market overreaction in the days prior to the announcements and a reversal effect occurring in the days after the announcement. The market overreaction hypothesis is yet another theory that contradicts Fama’s (1965) Efficient Market Hypothesis and is a key topic within behavioural finance. According to Nam, Pyun, & Avard (2001) the asymmetry is due to the mispricing behavior on the part of investors who overreact to certain market news. Research by Chopra, Lakonishok, & Ritter (1992) also found returns consistent with the overeaction hypothesis from observing short windows around quarterly earnings announcements. This analysis however was conducted under real world market conditions and doesn’t take in to account that virtual markets are open 24/7.

From the analysis of the existing cases prior, it would appear that the information announcements act as a market anomaly on its own. It appears that market overreaction triggers an investor to hedge or speculate and as a result massively increases the volatility of a share price. This volatility effectively allows abnormal returns to be gained which was a major critique to the efficient market hypothesis. Market anomalies are discussed in further detail in the next section of this chapter.

2.4 Anomalies in Capital Markets

Many observed market movements are not explained by the arguments presented by the efficient market hypothesis contested earlier. In the standard finance theory, market movements that are inconsistent with the efficient market hypothesis are called anomalies (Bostanci, 2003). According to Tversky and Kahneman (1986) “an anomaly is a deviation from the presently accepted paradigms that is too widespread to be ignored, too systematic to be dismissed as random error, and too fundamental to be accommodated by relaxing the normative system” (p. 252). Examples of Anomalies in capital markets are; January Effect, Low Book Value Effect, Neglected Firm Effect, the Reversal Effect, Days of the Week Effect, Dogs of the Dow Effect and the Intra Day Effect. It is also worth mentioning the potential information announcement anomaly discussed in the previous section of this chapter as it does play a role in achieving abnormal returns in spite of the efficient market hypothesis.

illustration not visible in this excerpt

Table 3: A Table to show a list of anomalies and their main contributors

2.5 Intra-Day Effect

The Intra-Day Effect or Time of The Day Effect is an anomaly found in financial markets, which is part of a large critique to Fama’s efficient market hypothesis described previously. Fama’s (1965; 1969; 1970) research denotes that if a market is Semi Strong Form efficient, that the market cannot be beaten unless insider trading occurs. However, findings from Wood et al. , (1985), Harris (1986) and Bildik (2010) has suggested that the intra-day effect does exist in capital markets thus between specific time intervals during the trading day, abnormal returns can be obtained. There has been a significant amount of research by different theorists regarding this anomaly as analysed below.

illustration not visible in this excerpt

Table 4: A Table to show the main theorists regarding the Intra-Day Effect

Robert A Wood, Thomas H McInish and Keith J Ord (1985)

Wood et al. (1985) were three of the first to analyse the anomaly ‘The Intra-Day Effect’ or ‘The Time of The Day Effect’ as it more commonly known. Their investigation into the nature of the return-generating process and the characteristics of trade size, price changes, trading frequency, and trading interval at the level of individual trades of stocks was derived from a large sample of the NYSE. Wood et al. (1985) found previous studies of market indices used daily, weekly, or monthly time-aggregated returns. Wood et al. (1985) used an equally weighted index of common stocks listed on the NYSE and examined two periods comprising of the six months from September 1971 to February 1972 and calendar year 1982.

Figure 2 below indicates that the Standard Deviation market return and risk are higher during the first few minutes of trading. For the period 1971-1972, the ratio of market return to its standard deviation, the reward/risk ratio, shows a negative trend from just over 0.3 near the opening to zero within 60 minutes. This suggests that there are early periods in the day in which abnormal returns can be obtained. This supports the intra-day effect anomaly and therefor supports the cause for undertaking the research in an in game virtual market environment.

illustration not visible in this excerpt

Figure 2: Standard Deviation of Market Return across Days (by minute, overnight returns included)

(Wood et al. , 1985)

Lawrence Harris (1986)

Harris (1986) set about investigating systematic weekday differences in open-to-close returns. Along with Wood et al. (1985) and Bildik (2001), they wanted to investigate the anomaly known as the Intraday or Time of the Day Effect. To investigate systematic weekday differences in open-to-close returns, Harris (1986) computed means by 15-minute intervals of the returns that accrue within the trading day. Cumulative means were plotted by weekday in Figure 3 below. There is an obvious difference in the returns in the opening 45 minutes between Monday and the rest of the days in the trading week. The mean return in the interval for the NYSE equal-weighted portfolio is negative on Monday (-0.13%), while on the other days of the week it is positive (0.09%, 0.14%, 0.12%, and 0.10%). The difference is significant.

The results indicate that there are systematic time-series patterns in mean intraday returns that are common to all of the weekdays. Even within weekday trading periods, prices do not evolve at equal rates. Like the results from Wood et al’s. , (1985) research, this research is evidence that there is an intra-day effect anomaly occurring in real world capital markets. Based on what this anomaly means, abnormal returns can be achieved through picking stocks at specific time intervals during the day.

illustration not visible in this excerpt

Figure 3 : Intra-daily Patterns in Stock Returns (Harris, 1986)

Recep Bildik (2001)

Bildik (2001) on the other hand calculated the mean intraday 15-minute returns from the 1 January 1996 to 1 January 1999. Figure 4 displays the movements of the 15-min returns throughout the day. It showed that there was a decreasing trend at the beginning of the day and also an upward movement in stock returns at the end of the trading day which represents typical U-shape or more precisely W-shape pattern in stock prices shown on Figure 4. Similarly, Wood et al. (1985) and Harris (1986) also found U-Shape/W-shape patterns in their empirical research when plotting the standard deviation and means on a graph. These results along with research by Wood et al. (1985) and Harris (1986) support the existence of the Intra-Day effect as an anomoly in capital markets. However despite these results, the research was not carried out under virtual market circumstances and therefor do not respresent what happens within them.

illustration not visible in this excerpt

Figure 4: The Mean Intra-day 15-minute returns (Bildik, 2001)

2.6 Judgments and Values in Computer Games

In order to perform analysis on efficient market hypothesis under virtual market conditions, there must first be a cross-sectioned analysis of the economics of virtual markets. This then provides a platform to compare and contrast with real market results and allow the differences to produce critiques to the methods posed by real world efficient market hypothesis research.

In-Game Capital

As Malaby (2006) said, “In-Game capital could be dissected into three distinct categories. The three categories consist of Material Capital, Social Capital, and Cultural Capital.” (p. 141) In-game capital is the accumulation of material, social and cultural capital that customers perceive to possess within their respective game. As Malaby (2006) described, this is solely a perceptual value, not an actual value. These capitals embody the switching costs customers will incur when they are no longer able to interact, use, or control the capital they have accumulated or generated because of the termination of the relationship.

According to Malaby (2006) “what the customer perceives the particular resource can be used for is the value of game capital.” (p. 146). Game capital is therefore not a static value. For example, high cultural capital may be perceived by customers as permitting them access to more content from their respective game. Game capital represents the investment a customer holds in the game. Each of the three in game capital forms combine that influences the perceived termination value of the relationship.

Virtual Consumerism

Lehdonvirta, Wilska, & Johnson (2009) said that for at least two decades, simulated shopping and commodity consumption have been central elements of play in many digital games and online hangouts. Players or participants, often young, are familiar with the logic of consumption and ownership and gladly engage in the simulated consumption games offered by the systems. In recent years, it has become increasingly common for virtual goods circulated in consumption games to be exchangeable for real money Lehdonvirta et al. , (2009).

Using a credit card, mobile phone or topped up gift cards, players are now able purchase virtual items, clothes and characters like any commodities in an online store except that the goods are delivered in game as opposed to the customer’s doorstep Lehdonvirta et al. , (2009). In the past five years companies such as Facebook who due to its popularity have generated significant revenues by selling virtual commodities to its users. Common commodities usually priced at a dollar or less and some more sought after commodities can reach up to tens of dollars. Microsoft’s XBOX has seen more than 2 million subscribers join the service since 2005 and World of War craft has seen more than 5 million subscribers join Herrera (2010). These subscribers are then more than likely to purchase in-game virtual commodities as they have already established the relationship needed by paying for the subscription to the service alone.

2.7 Common Exchange Mechanisms

As the stock exchange provides the medium for traders to buy and sell stocks in a real world scenario, the equivalent of this has been developed within a virtual market in the following table. The Exchange mechanisms from within virtual markets are presented in a rough chronological order in which they first appeared in virtual economies. The order also reflects the development of commerce in general economic history.

illustration not visible in this excerpt

Table 5: Ten Exchange Mechanisms that implement markets (Lehdonvirta & Castronova 2014, p. 128)

Auction and Buyout House

Old-fashioned auction houses have been replaced in modern times with digital equivalents in both national and virtual economies. According to Lehdonvirta & Castronova (2014), MMO game World of Warcraft set the standard for in game auction houses with their highly efficient system. Players drag their items onto the trade screen and set a starting bid and length of time for the auction to last. Other players can then search for the item and register bids before the final countdown. The winner receives the item via the in game mail service and the seller receives the price less a transaction cost via the same system. All other bidders are returned their bids by the same system also. The auction house acts as an escrow for buyer’s further improving efficiency in the market.

Buyout Houses are simply auction houses with the added functionality of setting a fixed price as well as a starting auction price. In 2009, Electronic Arts Inc. introduced a new game mode on its FIFA titles. The new game mode is called Ultimate Team and it consists of users buying player packs to build the ultimate football team to compete against other players and accumulate coins in the process. The packs can be bought for either real money, FIFA Coins and more recently, FIFA Points. The players are presented as cards and can be traded on FIFA’s in game auction house that has the Buy It Now functionality. Users can search for players via formation, price, name attributes, and positions. Like World of Warcraft, The buyout house acts as an escrow and safely transfers the price minus a transaction fee and the card to the new owners. Despite users being able to purchase FIFA Points for real money, capital is retained within the game referring back to In Game Capital in 8.6 Judgements and Values in Computer Games. It must also be pointed out that unlike real world markets, virtual markets do not close, and it is always peak trading time for someone somewhere in the world.

2.8 Regulation of Virtual Markets

In real world markets traders pay premiums on top of their sales and purchases of stocks to help with maintaining a fair exchange. The same applies to virtual markets in the sense that a user will receive their item or currency plus or minus a small percentage added or removed to make it harder to obtain abnormal returns. Efficiency and Regulation go hand in hand in virtual economies. Lehdonvirta & Castronova (2014, p. 129) said an overly efficient market would allow users to collect items and increase levels very easily and thus reduce the game publisher’s potential earnings.

An inefficient market will frustrate players and block any reasonable progress by the player within the game, thus once again reducing the game publisher’s potential earnings. Retiring virtual currencies and applying taxes to virtual markets helps to reduce the amount of purchasing power that the user may have. Users with lots of purchasing power can control the market and buy up large amounts of specific cards or commodities in order to arbitrage and force the prices up due to a reduced influx of the card or commodity available to everyone else. The simple supply and demand model supports this.

2.9 Chapter Summary

To summarise, Efficient Market Hypothesis and Random Walk theory have been widely studied and accepted despite its critiques for hundreds of years. Information announcements in a semi strong form efficient immediately affect the share price of a stock as it incorporates the new information into its price. Market overreaction affects the volatility of market prices in wake of the information announcement. This was shown with the cases shown in Table 2. Market Anomalies such as the intra-day effect have been proven to exist in real world markets therefor contradicting efficient market hypothesis. In-game Markets are built and regulated in similar methods to real world markets in the sense that they are regulated and controlled with auction and buyout houses acting as broker mediums however, virtual markets operate 24/7 as opposed to the opening and closing times of stock exchanges.

Chapter 3 - Methodology

3.1 Introduction

This chapter discusses the methodology that is used in this dissertation. The first section describes the research questions asked and the hypothesis tests uses if any. The second part describes the method in which the data for the research will be collected. Subsequent sections include the sampling techniques used to achieve accurate results and the methods of data analysis along with its limitations and validity. This research study was conducted based on the methodology. This methodology plays an important role in implementing this research study accordingly. The details of the methodology are explained in detail in this chapter.

3.2 Research Questions

The questions we want to find out are; “Do in-game virtual markets suffer from the Intra-Day Effect? “What happens when new information becomes publicly available?” and “Does this anomaly reflect what happens in real world markets?” As previously mentioned in the main introduction, there is no research currently being undertaken on this subject area within an in game virtual marketplace. The results of this research could help game designers to eliminate unfair advantages within their games’ economies. Our chosen in-game market is on the XBOX 360 version of Electronic Arts’ FIFA Ultimate Team 2015. How these questions will be answered is tackled later in this chapter.

3.3 Quota Sampling

Before any research can be collected, for purposes of accuracy, there must be a sample size generated which will reflect the total population of the market. In this case, it was identified that there were 11,625 individual ‘player cards’ as at 05/01/15 (We Fut, 2015). The sampling technique used was a quota sample. This was used because the sample quota could be disaggregated into sub-categories (Saunders, 2009). The subcategories consisted of a position undertaken on the football pitch. These included Goal Keeper, Centre Back, Left Back, Right Back, Centre Defensive Midfield, Centre Midfield, Right Midfield, Left Midfield, Centre Forward, and Striker. Amongst these categories was a hierarchy of Gold, Silver and Bronze status in correlation to the players’ real world ability. Table 6 below shows the breakdown of the sample. The quota sample percentage used was 0.5%. Where the quota fell below one, a quota of one was used and all quota figures were rounded up to the nearest whole number.

[...]