Cover me, I'm going public! : “The relationship between IPO and media coverage”

(1)

Cover me, I’m going

public!

“The relationship between IPO and media coverage”

MASTER THESIS STUDY THESIS WITHIN: Finance NUMBER OF CREDITS: 30

PROGRAMME OF STUDY: Civilekonom AUTHOR: Carl Strand & Ronni Samir Sulaka SUPERVISOR: Fredrik Hansen

(2)

ii

Acknowledgment

Four years in Jönköping is now coming to an end. This thesis is written in the spring of 2018 and marks the end of our studies at the Civilekonom Program, Jönköping International Business School. We hope this essay will be of interest to the readers and inspires further research. We are very thankful to our supervisor Fredrik Hansen for his guidance and help during this work. In addition, our sincere appreciations to every objector and a special thanks to Aleksandar Petreski that has reviewed the statistical parts of this study and shared his thoughts with us.

Jönköping, 21 May 2018

(3)

iii

Summary

Background: The number of IPOs has been numerous in recent years. Moreover, the returns of the IPOs managed to either be over and underpriced. IPO-firms are generally unknown before they get listed. Therefore, the media outlets play an essential role in the dissemination of information to new investors. Thus, it becomes noteworthy to investigate whether media attention could be an explanatory factor for the first-day returns and how it affects an IPO in the short-term perspective.

Research Questions: i) If the amount of media coverage two months pre-IPO has any

relation with the first trading day return and ii) If the amount of media coverage helps to predict the IPOs stock return volatility after a two-month period (44 trading days)?

Purpose: This research will attempt to find evidence if the amount of media coverage pre-IPO may drive the demand for the pre-IPO and the first-day return. For this purpose, it is also necessary to find the under and/or overpricing. Furthermore, a regression analysis will be applied during a two-month period after the first trading day to investigate if increasing volatility depends on the amount of media coverage.

Delimitations: The sample consists of 165 IPOs in Sweden from Aktietorget and OMX Stockholm during the period 2005-2017. IPOs are initial introductions that are not unit IPOs, mergers & acquisitions, right issues, spin-offs or buy-out firms.

Method: The first research question is explained by a Pearson correlation where X is the media variable and tested against the degree of under or overpricing. Furthermore, a multiple-linear regression is examined where variables market index, Retriever data and trading volume is tested against stock return.

Conclusion: It has been identified that Aktietorget has been overpriced 2005-2017 by 2.9% while OMXS has been underpriced 6.8%. In summary, the study did not manage statistically to ensure that the amount of media coverage significantly influenced the stock return volatility.

(4)

iv

Definitions

IPO - An initial public offering means that a firm chooses to take stock on the stock exchange. In connection with this, the public is often offered to subscribe/buy shares in the company to a specific subscription order.

Hot/Cold Market - When there is a high density of IPOs it is called a hot market and when the density is low it is a cold market. This is also due to the different interest rates set by the central bank.

Investment bank - An investment bank is a securities company that transfers and issues financial assets.

Money left on the table - Lost capital for the company due to underpricing. The IPO prospectus price could have been valued higher to raise more equity to the firm.

Roadshow - Refers to when managers of the company going public try to attract potential investor through a presentation.

Prospectus - A prospectus is a financial document provided by the issuing firm and the investment bank. The prospectus contains the information required for an investor to make an assessment of the risk and return associated with an investment in the financial instrument.

Lead underwriters - Is referred to the investment bank that organizes the IPO process.

Book-building - The process which the underwriters of the IPOs try to determine the share price.

POP - Public offering price is the price set by the underwriters, this is the IPO stock price offered to the public.

IFRS - IFRS or International Financial Reporting Standards is an international regulatory framework that concern all listed companies.

Crowdfunding - Method of financing projects or ideas through Internet-based systems. This often concerns minor and new business concepts.

(8)

1

1. Introduction

In this chapter, the reader will be introduced to the subject with a brief background on the mechanism behind the initial public offering to enter the market. Also, the research question will be presented and the delimitations of this thesis.

1.1 Background

The recent year's low-interest-rate environment has contributed to records in new listings on the Swedish Stock Exchange (NASDAQ, 2017). The companies are, in most cases, relatively unknown to the capital market in which they seek financing (Friedlan, 1994). Therefore, media's role as an intermediate is particularly important as a distributor of information to the public to raise the interest from the potential investors. Shiller (2000) examined the IT bubble during the millennium where he discovered that media-hyped Internet stocks bear some of the responsibility for the Internet bubble. Further, Prechter (2001) argues that numerous people depend on following the flock and the information from the media. Individuals consider themselves to have a knowledge gap and do not possess enough knowledge within the complex world of investments (Prechter, 2001).

A common denominator for these new listings has, therefore, meant that most stocks close at a price higher than the offering price given in the prospectus pre-IPO - a phenomenon called “underpricing” (Ibbotson and Ritter, 1994). Liu, Sherman and Zhang (2007) documented that more media coverage before the IPO day significantly related to larger IPO underpricing. Furthermore, media coverage also increases perceptions of value and thereby increases the demand of a firm's stock, driving its price up and increasing underpricing (Pollock and Rindova, 2003).

There are many factors today on what affects the price of an IPO such as how investors, investment bankers, firm’s managers, founders and owners process the information upon an IPO (Jenkinson and Ljungqvist, 2001). In the process of going public, IPO firms experience a vivid change in visibility and that is primarily determined by the media coverage (Bajo and Raimando, 2017).

Studies typically investigate factors that are at a firm level and scholars have therefore started to explore how media coverage change the influence of an IPO (Guldiken et al., 2017; Scheufele, Haas and Brosius, 2011; Bhattacharya et al., 2007).

(9)

2

IPO media coverage also affects the stock’s value, liquidity, analyst coverage, institutional investor ownership in long-term (Merton, 1987). All things combined make this framework interesting for investigation because the role of the media may have a significant impact on the average return on IPOs, since all IPO firms do not receive equal attention from the media (Friedlan, 1994).

1.2 Problem

Media exposure becomes highly relevant in the process of an IPO. Investors perceive information about new companies that in general are relatively unknown pre-IPO. Therefore, the media's role in IPO coverage becomes vital for investors decision making (Friedlan, 1994). Robert Shiller (2000) claimed that media-hyped Internet stocks and the hype were a major factor responsible for the Internet bubble. In contrary, Bhattacharya et al. (2007) rejected Shiller’s hypothesis and found no significant correlation between media exposure and IPOs. The lack of conclusions regarding the impact of media on IPOs is therefore ambiguous. The effect of a company associated with high concentration of media coverage in the start of the initial public offering benefits in the long-term since they are in the periphery of the investors (Merton, 1987).

Studies of similar character find that media news sentiment affect the investor’s decision which in turn influence stock returns and trading volumes (Tetlock, 2007). Furthermore, Scheufele, Haas and Brosius, (2011) also point out that first-trading prices deviate from the prospectus price - a phenomenon called under/over pricing. However, studies that examine the correlation between media coverage and the under/overpricing come from Germany and the American stock exchange. For the Swedish market there is a knowledge gap within this area which makes this research distinguished from similar studies.

Therefore, the results from this research can be used to understand whether an individual should invest or not in a certain IPO, depending on how much it has been covered in the media. The Pearson correlation will investigate if any predictions could be made pre-IPO to find the first-trading price depending on the sum of the number of media articles two months before. Further, a multiple-linear regression will be conducted to get an increased explanation on how much attention a given stock receives may thrive the stock movement.

(10)

3

1.3 Purpose

This research will attempt to find evidence if the amount of media coverage pre-IPO may drive the demand for the IPO and the first-day return. For this purpose, it is also necessary to find the under and/or overpricing. Furthermore, a regression analysis will be applied during a two-month period after the first trading day to investigate if increasing volatility relates to the amount of media coverage.

1.4 Delimitations

The research will have Swedish market as a geographical limitation. The sample contained 360 IPOs including First North, NASDAQ OMX Stockholm (OMXS), Aktietorget and Nordic Growth Market (NGM). Due to missing data from First North and NGM, the data got reduced to a sample of 165 IPOs in the period 2005-2017 and contained only Aktietorget and OMXS.

In line with prior IPO research of Scheufele, Haas and Brosius (2011) that investigates the short-term relationship between media coverage and stock prices, they chose to define a two month period as the short-term dimension in their study. The reason for a time period of 2005-2017 is that the appropriate data was not available before 2005, especially concerning prospectus prices. The measurement period of 12 years is considered to be representative to ensure the results of the study statistically. The interval of this study also contains two major stock market crashes, which makes the chosen data even more interesting since media could have even a major impact on stock prices.

One natural question is whether the amount of media coverage will be less relevant if it is not counted for negative or positive news. The argument from Liu, Sherman and Zhang (2007) is that it affects an investor’s opportunity cost of evaluating and hence may influence underpricing. Furthermore, they also argue whether if media coverage is bad or good, it is still associated with an asymmetric impact on stock price. Similarly, Barber and Odean (2008) argue that investors have limited time and that attention is a scarce resource. Options that attract attention are more likely to be chosen, regarding if it is negative or positive article (Barber and Odean, 2008). On the contrary, Bajo and Raimando (2017) find that positive sentiment has an effect. However, due to time constrain the sentiment of the media will not be considered.

(11)

4

An essential key to the research will be to collect the amount of media articles data from Retriever Analysis. The research will cover Swedish media outlets and Retriever will therefore collect data from 3736 Swedish media sources.

In line with Bajo and Raimando (2017), this research will also exclude unit IPOs, right issues, spin-offs, foreign firms, buy-out firms. New introductions are initial introductions that are not unit IPOs, name changes, mergers & acquisitions and foreign companies. Companies that have been abolished since initial introduction are also not included due to lack of information about these firms.

1.5 Research Questions

The questions that will be investigated in the following study are;

-If the amount of media coverage two months pre-IPO has any relation to the first trading day return?

-If the amount of media coverage helps to predict the IPOs stock return volatility in the

short term - more precisely after a two-month period (44 trading days)?

The degree of under or overpricing will be needed to answer the first question. Furthermore, the rate environment in different periods that is mentioned in the theoretical framework will be discussed to explain why managers choose to go public and how the market reacts on IPOs in different yield cycles.

(12)

5

2. Theoretical framework

In this section, the research will present and summarize all the theoretical framework relevant to the research subject. A brief explanation of the step behind the IPO process, market timing, the efficient market hypothesis, information asymmetry and the importance of behavioral finance when there is missing information about securities.

2.1 The IPO process

There are several reasons for a company to go public. A private company wishes to raise capital from a broader spectrum of investors and they proceed by new shares with help from investment bankers (Bodie, Kane, Marcus, 2013). Initial Public Offering (IPO) is referred to the first issue of shares to the general public. The IPO process involves the issuing firm, the lead underwriter(s) which are an investment bank and the private investors (Jenkinson and Ljungqvist, 2001).

Before even considering going public the company has to analyze the benefits and possibilities of a listing (Jenkinson and Ljungqvist, 2001). Jenkinson and Ljungqvist (2001) discuss the most common motives why companies desire to go public. Their research validates that a reason for going public could concern raising new capital and also achieve a more diversified group of investors for the sake of liquidity. Another reason is also to use the shares for mergers and acquisitions. Further reasons may be that an IPOs signals stability and reliability to customers and suppliers (Jenkinson and Ljungqvist, 2001).

Rydqvist (1997) found evidence that a portion of the offerings before 1990 was motivated by companies to compensate the employees by allocating shares upon on an IPO. Rydqvist (1997) found a connection between the high marginal tax rates on labor income and the lower tax rates on capital gains. Furthermore, the hypothesis was also supported that the company issued securities directly to the employees and that they were favored in the prospectus.

The first step of going public is to make sure that the company can satisfy the regulations set by NASDAQ OMX Stockholm (Sweden's Financial Service Authority, 2016).

(13)

6

This process includes two crucial separate procedures, finding investors willing to purchase shares and the company’s share have to be admitted to stock exchange (Jenkinson and Ljungqvist, 2001).

The next phase is called initial information gathering, during which the lead underwriter’s primary focus is to work closely with the firm that is going public to carry out due diligence investigations. Beyond this, it is vital to produce the information required to satisfy the Financial Services Authority’s regulations, also called the prospectus (Jenkinson and Ljungqvist, 2001). A prospectus contains the necessary information for the potential investor decided whether or not to subscribe shares for the IPO (Jenkinson and Ljungqvist, 2001). The information could be such as future growth, economic condition, pricing and risks taken with the investment. The prospectus document in Sweden has to be approved by the Financial Services Authority before releasing it to public investors.

After producing a prospectus, final or preliminary, the next main stage is to market issues to investors; this could range from formal meetings to overseas meetings over the phone. Companies are advised by the investment bankers to undertake “road shows,” where the senior managers are trying to entice investors by presentations in some various locations, favorably where the density of institutional investors is concentrated (Jenkinson and Ljungqvist, 2001).

In the later stage it is crucial that independent market analysts do a valuation of the firm, for this process to be successful they have to have full access to the information flow as published in the prospectus (Jenkinson and Ljungqvist, 2001).

A vital issue is to value the public offering price (POP), lead managers use different valuation techniques to get a fair issue price. The underwriter also looks at factors such as the profitability, public trends, growth rates and even investors’ confidence so that the POP becomes attractive enough. When the share price has been set remains how the shares should be allocated amongst potential investors. This is also referred to book-building, the investor bid for the share and request a certain amount (Jenkinson and Ljungqvist, 2001). Liu, Sherman and Zhang (2007) show in their report that once the step after the initial price is set, is the moment when pre-IPO media coverage appear to reach out to potential private investors (Sweden's Financial Service Authority, 2016).

(14)

7

After the shares have been allocated amongst institutional investors, the public investors that have subscribed for shares will know if they will receive any shares or not, depending on how many people subscribing for the stock (Sweden's Financial Service Authority, 2007).

Figure 1: IPO-process described (Liu, Sherman and Zhang, 2007).

A Direct Public Offering (DPO) is an alternative way to go public. Unlike an IPO, a company that chose a DPO raise capital directly without an underwriter. A DPO could be done publicly in the marketplace for stocks but also in crowdfunding platforms. A company that chooses to have a DPO may have a bank, but they do not guarantee full subscription in the DPO as it could be the case in the IPO. Therefore, the DPO-firm jeopardies to receive less attention from financial institutions like banks and venture capital firms (Castillo, 2018). The benefits of a DPO is that the filing process can be done more economical advantageously and rapidly (Giddings and Everett, 1998). This procedure also has its negative consequences since the company must reach out to financiers without the assistance of an underwriter. However, a DPO may give a fair stock price in the market with fewer speculations in comparison with an IPO where the over/under pricing becomes a fact.

2.2 Swedish Index, Retriever Media Archive and the different marketplaces in Sweden

What marketplace a company chooses to be listed on depends on the size of the company, the motivation for the introduction and the type of investor the company wishes to attract.

No matter what marketplace is chosen, higher demands are imposed on companies whose shares are traded on a regulated and public markets.

(15)

8

Some regulations could be that financial statements must be in accordance with IFRS, custodial information must be public and disclosure of changes in shares and votes must be reported to Financial Advisory in Sweden (Sweden's Financial Service Authority, 2016).

2.2.1 Swedish Index

Index weights several shares and can, therefore, be used as a benchmark. The most widely used index in Sweden is the OMXSPI. It weighs the value of all the shares listed on the Stockholm Stock Exchange. Thus it shows an overall indication of the stock market development (NASDAQ OMX, 2018). Therefore, OMXSPI will be the corresponding index as a benchmark when performance is calculated in this research.

2.2.2 NASDAQ OMXS

NASDAQ OMX Stockholm is owned by NASDAQ OMX Group and is a global player that owns and operates market exchanges in the US, Nordic, Baltic and Dubai markets. For a listing on NASDAQ OMX Stockholm, the company requires at least 25 percent of the shares in widespread ownership. The company should also have a market value of at least one million euros for a listing to be conducted (NASDAQ OMX, 2018). Furthermore, the market value is decisive for which segment the company is ending. At NASDAQ OMX Stockholm, the companies are divided into three different segments; Large Cap, Mid Cap (Medium Company) and Small Cap (Small Company). Companies included in the Large Cap segment have a market value of more than one billion euros. Small Cap consists of companies whose market value is less than 150 million euros. Consequently, the Mid Cap segment consists of companies that hold a market value of at least EUR 150 million, but not over EUR 1 billion. (NASDAQ OMX, 2018).

2.2.3 Aktietorget

Aktietorget is primarily aimed at entrepreneurial companies that are in a growth phase (Aktietorget, 2018). Criteria for companies to be listed at Aktietorget is to have a spread among investors. At least 200 shareholders and ten percent of the capital in widespread ownership (Aktietorget, 2018).

(16)

9

Low liquidity characterizes Aktietorget in comparison with OMXS. Therefore, it will be valuable to compare these two lists because the company sizes differ and the media could have a disparate impact.

2.2.4 Retriever Media Analysis

The Retriever Media Archive is the Nordic region's most extensive digital news archive containing printed newspapers, magazines and business press (Retriever, 2018). There is a function in the database of Retriever that can search articles and analyze the collected data from all national newspapers, local newspapers and hundreds of magazines. There are 3736 Swedish media sources in total that Retriever analyzes. This includes the well-known financial press as Realtid, VaFinance, Dagens Industri, Morningstar and also non-financial newspapers like Aftonbladet, Expressen and JNytt. Retriever Media archive also includes financial media distributors like FinWire and Cision (Retriever, 2018). Cision and Finwire have collaborations with media distributors and exclusively attracts an audience that has an investor behavior (Cision, 2018). For instance, Cision is one of the world's leading players in the distribution and dissemination of press releases and other market information (Cision, 2018). Another media channel is FinWire that writes short news agency texts, but also interviews and more comprehensive texts. Finwire news is spread to all newspapers and news agencies. The news can be viewed on SvD, DI and via several banks websites through their distribution channel. The financial press and other media do not have the resources to monitor all listed companies. As a result, smaller and young companies generally receive less media surveillance than large-scale companies (Finwire, 2018). Therefore, FinWire specializing in monitoring small listed companies. It increases the company's visibility and increases the likelihood of further spread, which may be of great value to the company (FinWire, 2018).

2.3 Market timing, hot and cold markets

The central banks around the world have been eutrophied capital into the monetary system the recent years and Sweden is not an exception (The Swedish Central Bank, 2016). The reason is to increase peoples consuming behavior to raise the inflation to the goal of 2% (Riksbanken, 2017). The low repo rates also mean that the banks lower the interest rate on the saving accounts since it is cheaper for the bank to borrow money from the Swedish central bank (Riksbanken, 2017).

(17)

10

Regarding the required return, people start considering other options than the saving account and thrives on putting their money in the stock market (Riksbanken, 2016). This creates a purchase pressure that thrives prices and valuations of the markets in general since there is more money on the supply side (SvD, 2016). The low-interest rate environment has made it lucrative for the managers to make new listings on the public market.

It is not necessary that the main trigger for an IPO be financially related, several studies have shown that market timing is equally important for managers. A firm's management analyzes if there are a bullish stock market and long-term financial yields before setting a date for the IPO (Batnini and Hammami, 2015). Based on Lowry (2003) it can be advantageously to launch an IPO when the market has a more optimist view. Conversely, during a recession, it can have an opposite effect. The external advantageous economic conditions influence managers to raise equity through new shares to the public stock exchange.

Schultz (2003) demonstrated that companies with urgent financial needs are more likely to go public during unfavorable markets condition. This research will also discuss some empirical data of “hot” and “cold” markets, also called “IPO Waves.” Hot market indicates strong waves are indicating a high concentration of IPO activity. Less intense waves indicate a weaker climate also referred to as cold market.

Furthermore, Alti (2005) provided evidence of the existing “IPO Waves.” Companies launched during hot market retained more financing in their cash flow compared to the IPOs launched during cold market conditions. When scheduling an IPO process, market timing plays a salient part (Baker and Wurgler, 2002).

2.4 Efficient market hypothesis (EMH)

Eugene Fama (1970) argues that the stock market's most important role is to reflect the market value of shares and to spread the risk among owners. The ideal condition for the market is that pricing of assets gives a right signal of fair value. This means that both companies and investors can make decisions on the assumption that the price reflects all information available. In other words, the market is efficient and no one can avail market imbalances achieve arbitrary gains.

(18)

11

Even though these assumptions do not have to reflect the reality, Eugene Fama (1970) argues that it not be necessary for the market to be still effective. Eugene Fama (1970) believes that the market can be considered effective even if a critical mass of investors has access to the information available. If all investors interpret the information differently, none of them will be able to use this information regularly better than other investors. This can also be described as some investors overreacting to available information while other investors are more pessimistic, and thus the market is leveling out and remains as effective. Eugene Fama (1970) defined the effective market hypothesis regarding the amount of information available on the market and then divided the spectrum into three different forms, strong form, semi-solid form and weak form (Fama, 1970). The perfect effective market represents the strong form where all information is reflected in the asset price. The information is worldwide from both private, such as inside information, and public sources, annual reports and historical data. In the second part of the spectrum, there is the hypothesis of the weak form, which means that all historical information is reflected in the market price, and thus the historical information cannot be used to surpass the market. The middle path that possibly represents reality is half-form where all public information along with historical data represents the effective market (Fama, 1970).

2.4.1 Criticism of the effective market hypothesis

The effective market hypothesis assumes that all parties are rational and similarly value the information. Malkiel (2003) has summarized the criticisms made by other researchers regarding the hypothesis of market efficiency. Many critics argue that the market is irrational and that the market cannot be completely effective. The majority of criticisms made against the market hypothesis mean that psychological factors have a major impact. For example, critics argue that the "internet bubble" that occurred in the early 2000s clearly illustrates an irrational market. However, Eugene Fama (1970) was aware that the strong form of market efficiency was not realistic, but it should be regarded as an ideal state in an efficient market and has now become a benchmark for assessing market efficiency. Malkiel (2003) points out that, even though he sees the market as irrational, it includes the "week effect" and other seasonal trends have shown that certain periods. For example, near the weekend and after the Christmas period in January have historically yielded higher returns than others.

(19)

12

Although many investors are irrational, Malkiel (2003) states that when the rational investors try to exploit these historical patterns, they will influence the market and shift the impact. By placing the positive effect, investors therefore try to exploit the market for their gain. The problem is, however, that the information is available to all investors and if too many try to outwit the market, the effect will be omitted.

2.4.2 Random walk

The theory about Random Walk is a follow up on Eugene Famas (1965, 1970) theory about EMH. Random Walk assumes that the market has no memory and therefore is not affected by historical events. According to the theory, the current price of a product should match the correct price and all available information should be included in the price today (Fama, 1965).

In order for a stock exchange to follow a random walk, the market should be effective. In this case, an effective market can be defined as a market with a large number of rational investors that have access to all available information to maximize profit. If actual prices and correct prices differ systematically, these intelligent market participants will use that information to act better on the market. However, since all logical investors try to take advantage of this knowledge, it tends to neutralize the effects, which means that the current price will move towards the correct price, thus fulfilling the requirement for a Random Walk. Since a prevailing price is equal to the correct price, the price of the stocks changes only when new information reaches the market. Thus, the likelihood of positive price change is as high as the likelihood of negative change in stocks (Fama, 1965).

An important assumption is that investor’s act directly as new information reaches the market and thus eliminates the profitability, which is also closely related to the strong form of EMH. The question remains here whether these investors have access to all media information regarding an IPO and whether they act rationally. However, Random Walk does not require the market to be effective with only rational investors, as EMH does.

(20)

13

2.5 Information asymmetry

When a company gets listed, this means that ownership and control of the company will be diversified. Therefore, the management and owners will have to deal with a broader audience. As a result, the management has the best knowledge of the company, while owners and potential investors are referred to the information that the company discloses when evaluating the company and evaluating its performance (Palepu and Healy, 2007).

The situation, when a party possesses relevant information that the counterparty lacks knowledge about, is called asymmetric information (Akerlof, 1970). George Akerlof (1970) describes information asymmetries with an analogy in the well-known lemons-example. Where a car seller has complete information about the asset, or in this case the car's value and quality. The buyer, on the other hand, has only the information about the asset that the seller has chosen to share. On the market, there are other assets for sale at the same price, but if the buyer cannot determine the quality of the asset, then this risk will pay as much for an asset of more inferior quality as for one of higher quality. This analogy is based on the assumption that the buyer's decision is based on market statistics and historical values for the asset's average prices. As higher quality assets cannot be distinguished from those with lower quality, no access will be appropriately valued. As George Akerlof (1970) writes, there is the risk of paying expensive for an "acid lemon" when the value cannot be determined in advance. After the transaction, the buyer tastes on the "lemon" and gets a sour aftertaste of the purchase when the value has been realized. The following examples can also describe information asymmetries in other markets such as the stock market.

Through the prospect, investors can share the information that the company is willing to share (Palepu and Healy, 2007). Stock introductions are considered to be able to show a higher degree of this information problem than companies that have been on the stock market for a more extended period (Roch, 1986) since the company is often unknown and has not yet been evaluated of the capital market (Friedlan, 1994). The prospectus presented is often the first contact between the company and the capital market. It is the only document distributed before the listing is completed (Friedlan, 1994), which increases the uncertainty and the risk of the skewed information distribution.

(21)

14

The company's vendors, who possess the most knowledge of the company, may have the incentive to maximize the payment received at the launch by acting opportunistically and utilizing their information acquisition to increase the benefit of investors, (Darrough and Rangan, 2005). The reason is that the company's management possesses knowledgeable information about the company's actual assets, competitiveness, potential prospects and similar factors. Corresponding to information that the potential investors - buyers - request as a basis for a reasonable analysis (Darrough and Rangan, 2005). In turn, the media outlets may influence the stock price even more since they can draw conclusions and thoughts concerning the prospectus when they do not have the full information as the company’s management (Roch, 1986).

2.6 Herding and behavioral finance perspective

Herding behavior means that an individual tends to make the same decision as the larger group. It is a phenomenon that is well known for humans and this occurs physically through a mental impulse of the individual when it sees how the flock acts. The human genetic composition enables us to act more quickly in emotionally depressed situations rather than rationally and this is noted in the financial markets as well (Qawi, 2010). Prechter (2001) believes that many individuals base their investment decisions on the information they find in newspapers, news programs and other media channels. He claims that private investors do not trust their own ability and rather choose to trust experts. Therefore, many people depend on following the flock, the information from the media, because they consider they do not possess enough knowledge in the investment field. Thus, there is a lack of knowledge and strong driving force to follow others who create these behaviors on the market (Prechter, 2001). Qawi (2010) believes that it is not only private smaller players who demonstrate herding behavior, but also "financial professionals" which is due to the human need to strive for consensus. People usually follow the flock instead of making their own decisions, which can be more rational, is based on the fear of being considered differently. Herding behavior creates trends in the market and the stronger the flocking behavior, the longer the trends last. This results in either a price bubble or a sharp decline in the market, which is a contradiction to the theory Random Walk, which argues that market movements are only random (Qawi, 2010).

(22)

15

There are many examples of how herding has created strong market movements and led to that financial bubbles broke out. The IT hysteria that lasted around 1999’s is a clear example of herding where many investors stopped reasoning and continued to buy unreasonably highly valued IT shares. The whole market was astonishingly IT focused and very few investors dared to oppose the big mass (Vivek, 2012).

2.6.1 Overconfidence

Overconfidence is a human behavior when people tend to overestimate their competence, beliefs and also the precision of future forecasts (Barberis and Thaler, 2003). Investor behavior, such as overconfidence leads to overvaluation of risky stocks. Therefore, the required risk premium becomes lower and the investor pays a higher price for the security (Adebambo and Yan, 2017).

Jlassi, Naui and Mansour (2014) conducted an international research about overconfidence and the effect on dynamic market volatility. They provided empirical evidence of overconfidence with different degrees among financial markets. Individuals psychological mindset dictates investors and most often have short-term reasoning. Overconfidence played its part during the financial crisis when short-term behavior mainly increased stock prices fluctuations (Jlassi, Naui and Mansour, 2014).

The endorsement effect is another interesting phenomenon, which explains how decisions by financial experts or just recommendation of a stock can impact the share price (Hirshleifer and Teoh, 2003). According to a research done by McGee (1997), IPO underwriters use salient investors that have an influential position to create a buzz around the issue, just like shopping malls can open “anchor” stores to attract other stores. Therefore, prominent investors could make the volume of media-provided information raise more interest and attention which in turn affect the investor preferences (Pollock and Rindova, 2003).

(23)

16

3 Literature Review

The following chapters will highlight relevant and current research within the subject area of media impact on IPOs. Studies from Bajo and Raimando (2017) will be highlighted and finally, previous mixed studies will be examined. Initially, studies about the phenomenon underpricing will be presented since it is relevant for the solution of the hypothesis.

3.1 Underpricing

When a company decides to launch the IPO process, it needs to be valued and a price of the shares needs to be determined. There is usually no previous market price for the shares, but the company has to find a valuation that is reasonable and which investors are willing to pay (Friedlan, 1994; Ljungqvist, Cornelli and Goldreich, 2005). Thus, there is an outcome that the price of the share is either adequately valued, under or overpriced. Underpricing means that there is a difference between the subscription price of the share and the closing price of the share the first day after the IPO (Ljungqvist, Cornelli and Goldreich, 2005). Underpricing implies that owners could have received more capital if the underpricing had not occurred. Instead, the pricing becomes an advantage for investors where they get heavily discounted shares that later on can be sold at the closing price, thus making a potential profit already the first day (Loughran and Ritter, 2002). Research shows that the most stock introductions have been underpriced but that there is a wide variation between different companies and countries (Jenkinson and Ljungqvist, 2001). Ritter and Welch (2002) did a study about the first-day closing prices over the period from 1980 to 2001 and found that the underpricing was 18.8% in the U.S market. Previous studies about underpricing provided a sample size of 374 IPOs in Sweden for the period 1980-2011 and found that the initial average return was 27.2% (Rydqvist, 1997; Schuster, 2003; De Ridder, 2015).

(24)

17

Diagram 1: Initial average return in different markets (Loughran, Ritter and Rydqvist, 1994). Loughran and Ritter (2002) examined 3025 IPOs during the period 1990-1998 and explained the main reason with IPO underpricing was an indirect compensation for underwriters. Similarly, Bradley and Jordan (2002) documented that IPO underwriters do not disclosure public information which entails that IPO underpricing and abnormal initial returns are intentional. Additionally, Loughran and Ritters (2002) find that IPOs significantly benefits venture capitalists and executives of the issuing companies.

Nevertheless, this research will investigate whether underpricing could have any relationship with the amount media coverage in prior-IPO date and short-term. Liu, Sherman and Zhang (2007) found a positive relationship between underpricing and pre-IPO media coverage. However, Staikouras and Tsatsanis (2004) examined UK Company’s first-day returns from IPOs and the media exposure prior to IPO date. Their findings show that the final week’s media exposure is negatively related to underpricing.

DuCharme et al. (2001) examined internet companies in the US and found strong evidence that underpricing increases with the extent of media coverage of the IPO firms seven days prior listing. On the contrary, Bhattacharya et al. (2007) have shown that there is no significant correlation between media exposure and IPOs. In conclusion, different research has different views of either no significant, negative and also positive correlation between media exposure and underpricing.

(25)

18

3.2 Introduction to the role of media

Cutler, Poterba, and Summers (1989) were one of the first empirical studies trying to find the link between media coverage and stock prices. They are explaining that a sizable portion of the movements in markets occurs in connection with major news releases. They started by analyzing the monthly stock returns and measures macroeconomic activities between 1926-1985 in the United States. However, it is unlikely to use this study as a framework for the following research since extreme events like World War 2 and political changes has been fundamental in their research.

Some believe that media's purpose is to produce entertainment and not information (Jensen, 1998) while other scholars argue that media has a role as an information intermediary (Bushee et al., 2010). As an argument, Jensen (1998) uses the old media channels as a framework while media as it is today is quite different (Bushee et al., 2010).

Some researchers have argued that the media affect perceptions of legitimacy (Pollock and Rindova, 2003). Media also impacts the ways future stakeholders interpret and evaluate information about firms by framing its descriptions of them in positive and negative terms (Pollock and Rindova, 2003). The media legitimates firms by directing public attention and thereby increasing public’s exposure to them.

The general idea of an initial public offering is to seek new capital but also new investors. Thus, to fulfill the initial public offering, it is necessary to be detectable to the public so that new investors can find information about the company and the IPO itself (Liu, Sherman and Zhang, 2007).

A prospectus and a proper valuation are therefore not enough - the company must also be visible to the public. A company that reaches out to investors and other stakeholders gets more opportunities to raise equity. Furthermore, increased visibility and media exposure lead to increased awareness of the company, which can benefit the company's marketing and business. This means that the IPO-firms actively needs to strategically pursuit media coverage (Pollock and Rindova, 2003).

(26)

19

3.3 Media Sentiment and IPOs

The study of Bajo and Raimando (2017) describes that media plays a pivotal role when investors take decisions upon IPOs. A large of a portion of information is processed but is relied on second hand and simplified news that does not come from the company itself. Similarly, the study from Pollock and Rindova (2003) also shows there are many sophisticated and skeptical buyers upon IPOs and found out that their result is media-provided rather than company-media-provided. Hence, media influence the investor behaviors systematically. The sentiment of the media shapes the belief of investors regarding the IPO-firm and may increase (decrease) the demand for the share and the first-day returns (loss). Bajo and Raimando (2017) had a hypothesis that there is an essential aspect of how media presents the information as positive or negative because it may cause an impact on the investor’s belief of the issuing IPO firm. Depending on the tone, it may shape investors belief of the company and in turn, increase the demand for the share and the first-day returns. Ceteris paribus, in cases when the newspapers had a negative tone about an IPO, it makes the first-day return decrease (Bajo and Raimando, 2017).

In consensus with Barber and Odean (2008), they argue that investors are net buyers of attention-grabbing news. This means that investors purchase stocks that have caught their attention and continuing by that investors are not sensitive whether it is negative or positive information. Furthermore, Seasholes and Wu (2007) also explain that individuals have search problems by the endless stock supply. Hence, when the different media outlets announce a new prospectus, this becomes in the ”consideration set” for that individual investor. According to behavioral theories, the event of when the media outlets share new information about a new company, it means it grabs attention which helps the individuals to narrow the universe of the stocks that needs to be searched (Seasholes and Wu, 2007).

Nevertheless, there is also sufficient evidence from finance studies (e.g., Chan, 2004; Patell and Wolfson, 1984) about investment recommendations published in financial media or negative/positive news about a given stock. Price fluctuations often occur immediately or one day after the news have been published. Recommendations from analyst, especially about small-cap companies, can lead to abnormal returns. This effect usually diminishes within a couple of trading days (Chan, 2004).

(27)

20

Bajo and Raimando's research was based on 2814 IPOs from the American stock exchange and over 27 000 newspaper articles in four hundred newspapers. They measured the tone of the news detecting degree of positiveness, negativeness and uncertain sentiment. Bajo and Raimando ran a search with all different keywords on each article. For instance, words like “great,” “strong,” “better” and “profitable” would be counted as positive. “Postponed,” “canceled,” “poor” and “bad” would be negative. Thus, an article of 100 words would count of 4% positiveness (4 negative words out of 100 words) and the same holds for the negativeness if the previously mentioned words would be a real example. The next step in their study will be to find a correlation if the positiveness (negativeness) counts for higher (lower) first-trading price of that examined company. One thing that could be questioned in their research is whether it is reliable to measure with this approach for 27 000 articles.

The results found evidence that positive tones are positively associated with IPO underpricing. This means a significant positive relation between positive sentiments and is increasing the degree of underpricing by 2.5%. However, the negative tone in the media news appeared not to demonstrate significant fluctuation in the price behavior on the first day of trading.

3.4 Media coverage, stock price and short-term relationship

The research conducted by Scheufele, Haas and Brosius (2011) aims to find a short-term relationship between stock prices, media coverage and trading volumes of eight listed German companies. Based on famous theories like the efficient market hypothesis and the random walk theories by Eugene Fama (1965), the authors try to identify media's role in how stock price fluctuates. Scheufele, Haas and Brosius (2011) also argue that behavioral finance researchers meet a large number of cases where stock prices deviate from fundamentally justified levels. According to behavioral finance, the aggregate group of private investors are not homogenous and do not decide in a fully rational manner whereas the efficient market hypothesis assumes homogeneous reactions (Scheufele, Haas and Brosius, 2011).

In comparison with Bajo and Raimando (2017), the results from Scheufele, Haas and Brosius (2011) provided evidence that media coverage only has a minor impact on stock prices and larger companies received more attention in the information flow. They found clear evidence that the media coverage rather reflects than shape the stock exchange.

(28)

21

Nevertheless, Liu, Sherman and Zhang (2007) documented that more media coverage before the IPO day significantly related to larger IPO underpricing. Hence, more media articles are in fact associated with higher underpricing. They argue that one extra piece of media coverage is associated with approximately two percentage points greater underpricing. In comparison with Bajo and Raimando (2017), they could not find any evidence regarding positive media coverage and IPO firm’s long-run underperformance. Nevertheless, their research computed the IPOs in the US market 1980-2004, which in total was 3627 IPOs. The average first-day return was around 19%. Remarkably, 38% of the sample accounts for internet and technology firms. An argument here is the study of Bhattacharya et al. (2007) that examined the IPOs from 1994 to 2000 that rejected the hypothesis of that media coverage was a significant factor associated with the Internet bubble. Liu, Sherman and Zhang (2007) decided therefore to do a regression where they excluded the IPOs launched during the internet bubble period to make sure the results are not solely driven by the internet bubble. Although, their regression result indicated that the sample of 1980-2004 period yielded similar results even though they included the Internet bubble.

3.5 Summary of similar findings

Guldiken et al. (2017) conducted research exploring how the media coverage impacted IPOs stock returns in the first week when they were publicly traded. The report showed evidence that credible financial media in fact influenced how well the stock performed. This could be connected with signaling theory, explaining that if credible financial media spoke positively about a particular stock, traders viewed it as the media had an information advantage. The results from Guldiken et al. (2017) provided empirical evidence of signal theory effect and how it influenced actor behaviors. An important factor to consider is the herding behavior of individual investors.

Previous studies have also focused on the positive or negative sentiment produced by media (Pollock and Rindova, 2003). Arguing that media can create a legitimation of an IPO because of the lack of transparency. Hence, the more positive or negative news about a specific IPO will affect an IPOs stock price when it is traded. Pollock and Rindova (2003) also provided evidence that media plays an important role in firm legitimation. Media affect the salience and the perceived value of the firms which can be an influential factor upon an IPO.

(29)

22

A somehow focused perspective is offered by Jens et al. (2006) who researched the role of media coverage by only studying the biotechnology IPOs in Australia. They collected data from 30 Australian biotechnology IPOs in the period of 1994-2004. In line with this research, they also measured just the number of media citations and not the sentiment of the media in the Australian publications. However, they could not find any significant results that media coverage influenced the first trading day. Further, Jens et al. (2006) results showed that more money is left on the table as the IPO firm receives extensive media coverage. Again, this validates the important role the media channels have during the IPO process.

Nevertheless, Bhattacharya et al. (2007) did a study between 1996-2000 in the US on 458 IPOs and 458 non-Internet IPOs where 171 488 news articles were examined and classified as good, neutral or bad news. The examined period is interesting because Internet stocks went up by 1000% from January 1994 through March 24, 2000 to crash 45% in the next following months (Bhattacharya et al., 2007). The study was a follow up on the hypothesis from Shiller (2000) that claimed i) that media-hyped Internet stocks, and ii) that the hype

was a major factor responsible for the Internet bubble. Bhattacharya et al. (2007)

documented that media hype is unable to explain the Internet bubble because only a 1,6% difference existed in returns between Internet stocks and non-Internet stocks, and that media coverage only explained 2,9% of that and therefore rejected the second hypothesis of Shiller.

(30)

23

4 Method

This section will explain the research method, the research approach and research design. A describing and narrative analyze how data was obtained and difficulties faced. This is followed by the quantitative data and statistical approaches. Furthermore, the last part discusses the validity and reliability of this thesis.

4.1 Research Approach

The process of analyzing and obtain data is referred as the method according to Lewis et al. (2016). This thesis will try to explain the relation between media coverage, first-day return IPOs and stock return fluctuation over a two month period in relation to how much attention a firm has received in the media. The research can be labeled as deductive research, examination and conclusion are based on previous facts and search for new evidence in empirical data.

This thesis will emphasize the positivism approach. The interpretation of positivism approach relies on the researchers to undertake a value-free way mindset. The researcher cannot change certain facts that are given and can’t alter the data nor have any emotional attachments to the subject. “The researcher is independent of and neither affects nor is

affected by the subject of the research” (Remenyi et al., 1998, pp. 162). In this research,

all data are given and secondary data is obtained from independent websites.

4.2 Research Design

What is happening; to seek new insights; to ask questions and to assess phenomena in a new light” (Robson, 2002, pp. 59). When conducting research using exploratory research,

it becomes more familiar with a problem and an area. The descriptive research design is used when explaining a specific situation (Malhotra and Grover, 1998). The exploratory approach is usually connected with quantitative data in order to explain/find a relationship between variables. As stated by Aliaga and Gunderson (2002) quantitative research is

“Quantitative research is explaining phenomena by collecting numerical data that are analyzed using mathematically based methods (in particular statistics).” Hence, this thesis

will emphasize the exploratory approach to find a relationship between the variables that are examined.

(31)

24

To fulfill the first purpose of this thesis; “If the amount of media coverage two months

pre-IPO has any relation with the first trading day return,“ the under and overpricing will be

calculated as a first step. Secondly, a Pearson correlation will be measured between first-trading day return among the Retriever articles two months pre-IPO date. A Pearson correlation measures the relationship between the variables Retriever and first-trading day. The Pearson coefficient ranges from +1 to -1, where +1 indicates a perfect positive relationship, 0 no relationship and -1 a perfect negative relationship (Hair et al. 2010).

The hypothesis of the results from Scheufele, Haas and Brosius (2011) found strong evidence that media coverage only has a minor impact on stock prices and larger companies received more attention in the information flow. Thus, separate tests will be done on Aktietorget and OMXS since those markets attract companies with different sizes and capital. Therefore, the Pearson correlation will be measured between the two different marketplaces.

Hair et al. (2010) mean that the sample size is one of the most influential elements under the control of the researcher in the part of research design. The size of the sample size is directly linked to the statistical power of significance and generalizability of the result. Furthermore, Hair et al. (2010) argue that a number of 100 observations in most research situations is adequate. The sample of this study consisted all available IPOs in OMXS and Aktietorget and included 165 IPOs which is above 100. Nevertheless, previous studies had a wide variety in their sample. In comparison, Bajo and Raimando (2017) had 2814 U.S IPOs and 27 000 articles while Scheufele, Haas and Brosius (2011) just had eight German companies.

4.2.1 Research Strategy

A qualitative method data collection is based on words and interpretive analysis. However, this study is looking for statistical evidence and therefore emphasize a quantitative approach (Aliaga and Gunderson, 2002). More specifically, use one of the most important tools of econometrics, linear regression model (Gujarati, 2015).

The purpose of the linear regression is to find empirical evidence that media influences stock price volatility.

(32)

25

This research will combine time series data and cross-sectional data, also called panel data. Time-series data is observations collected over a period, for example, daily stock returns (Gujarati, 2015). Cross-sectional data is when variables are collected at a single point in time (Brooks, 2008). Since IPOs have different launching dates, this study combines both data techniques. As stated in 1.3 Purpose, this thesis examines how media influence stock volatility over a two month period, therefore a linear regression is compatible to use. Since there is no stock price data available before going public, the regression only includes the two first month of trading.

4.3 Data collection

In order to conduct the research, the daily stock price, daily index return and daily volume data were obtained from Datastream. Media articles were obtained from Retriever. The prospectus price was found in Prospectus document and from the Swedish Tax Agency database. Data downloaded from a trustworthy database is referred to secondary data and the main characteristics for this type of data are that it is collected by someone else than the end user (Lewis et al., 2016). For an individual researcher, it can be time-consuming and cumbersome to collect primary data. Therefore, using secondary data is an advantage to conduct a thesis when the time is of the essence (Lewis et al., 2016).

This study started naturally by finding all the IPOs that have been listed on NASDAQ OMX Stockholm, NASDAQ First North, Aktietorget and Nordic Growth Market. After reviewing all the data needed, the decision was made to investigate only OMXS and Aktietorget. OMXS and Aktietorget were the only two listings which had all data necessary to conduct this statistical tests. The study is based on data between 2005 and 2017 involving 165 IPOs from two different listings in Sweden.

In order not to get skewed representation, data that occurred during the financial crisis has also been included. An argument here could be that media had a negative sentiment during this dismay as examined in the study of Bajo and Raimando (2017). In comparison with Bajo And Raimando (2017), the purpose of this research states “if the volatility depends on the amount of media coverage,” making fluctuations even more interesting to study in times of financial crises.

(33)

26 4.3.1 Construction of media coverage variable

Every search is related to each collected IPO in the Swedish newspaper collections, provided by Retriever Database. Retriever includes the cast majority of the Swedish newspapers, using the firm name as the keyword. Just in line with Jens et al. (2006), the keyword was used as ‘<company name>’ without company form as “AB,” as in Aktiebolag. In line with Scheufele, Haas and Brosius (2011), the news search range spans from two months before and two months after the first-day of trading. In their research, they define this as short-time for the purpose of their research. While running the search on Retriever Database of the company names as search criterion dataset of 55205 articles ware generated from 3736 different outlets covering 165 IPOs.

4.4 Pearson correlation and over/underpricing

A list of all the IPOs is trading in OMXS and Aktietorget whose first-trading day between 2005-2017 was gathered and joined along with the media mentions from Retriever Database. Secondly, the daily closing prices for the given stocks starting from the first trading day was computed aligned with the media coverage and the stock prices two months before. The public offering price (POP) was gathered from the Swedish Tax Agency which is a reliable source but was still controlled along with the prospectus document from the given IPOs to double-check credibility.

The degree of under or overpricing will be counted as a first step to find a relationship with the number of media mentions in order to answer “If the amount of media coverage

two months pre-IPO have any relation with the first trading day return?” Under/overpricing formula:

(Closing price−Offering price) 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝑝𝑝𝑂𝑂𝑂𝑂𝑝𝑝𝑂𝑂

Pearson correlation formula:

(34)

27

Where, x equals the amount of media coverage two months ahead of an IPO, x̄ equals the average media coverage two months pre-IPO in Aktietorget and OMXS separately.

Where, y equals percentages initial return for given stock, Ȳ equals the average percentages initial return in Aktietorget and OMXS separately.

4.5 Multiple-linear regression hypothesis

The following research hypothesis has been developed:

𝑯𝑯_{𝟎𝟎,𝟐𝟐}: Media mentions have no significant impact on the IPOs stock volatility in short term

(44 trading days)

𝑯𝑯𝟎𝟎,𝟐𝟐: Media mentions have a significant impact on the IPOs stock volatility in short term

(44 trading days)

4.5.1Multiple-linear regression model (2 months after IPO)

Cutler, Poterba and Summers (1989) explain that a large portion of the market movements occurs in connection with major news releases. To describe and evaluate how different variables are moving in relation to each other, multiple-linear regression is the optimal tool to use (Brooks, 2008).

To evaluate a data set with regression one has to identify the dependent variable and the independent variables. Is it possible that the movement in the dependent variable Y can be explained by the independent 𝑋𝑋₁, 𝑋𝑋₂, 𝑋𝑋₃ (Gujarati and Porter, 2009). The independent variables that are included in the regression are the market index, Retriever data and volume. All the independent variables are daily data and the dependent variable is the daily closing stock return. Hence, the regression measures how the stock volatility is affected by the independent variables. The central limit theorem states that a sample over 30 observations is sufficient enough and the assumptions hold (Buglear, 2012).

(35)

28

The justification for the selected variables come from analyzing what affects the stock return and what previous literature has done before. The stock return is the dependent variable and since it could either be positive or negative return the research question two has been defined this as “volatility.” Further, the independent variable volume turnover indicates the market demand for a stock and for IPOs it gives an idea if a stock is over or undersubscribed (Pollock and Rindova, 2003). The market index indicates the stock market development and is therefore a relevant variable to test against the IPOs stock returns (NASDAQ OMX, 2018). Finally, the Retriever variable will measure the amount of media articles and is the most important variable in order to answer the research question, see Appendix J for example.

Hence, this study aims to find a relation between stock return Y, and the independent variables Retriever𝑋𝑋₁, market index return 𝑋𝑋₂, trading volume is 𝑋𝑋₃ while 𝑋𝑋₄, 𝑋𝑋₅ and 𝑋𝑋₆ are dummy variables tested. Hence, there will be the following equations:

𝑌𝑌_{𝑠𝑠𝑠𝑠𝑠𝑠𝑝𝑝𝑠𝑠𝑂𝑂𝑂𝑂𝑠𝑠𝑠𝑠𝑂𝑂𝑂𝑂} = 𝛽𝛽₀+ 𝛽𝛽_𝑥𝑥+ 𝜀𝜀 (Equation 1) 𝑌𝑌𝑠𝑠𝑠𝑠𝑠𝑠𝑝𝑝𝑠𝑠𝑂𝑂𝑂𝑂𝑠𝑠𝑠𝑠𝑂𝑂𝑂𝑂 = 𝛽𝛽0+ 𝛽𝛽𝑚𝑚𝑚𝑚𝑂𝑂𝑠𝑠𝑂𝑂𝑠𝑠 𝑂𝑂𝑂𝑂𝑖𝑖𝑂𝑂𝑥𝑥+ 𝜀𝜀 (Equation 2) 𝑌𝑌_{𝑠𝑠𝑠𝑠𝑠𝑠𝑝𝑝𝑠𝑠𝑂𝑂𝑂𝑂𝑠𝑠𝑠𝑠𝑂𝑂𝑂𝑂} = 𝛽𝛽₀+ 𝛽𝛽_{𝑂𝑂𝑂𝑂𝑠𝑠𝑂𝑂𝑂𝑂𝑂𝑂𝑟𝑟𝑂𝑂𝑂𝑂}+ 𝜀𝜀 (Equation 3) 𝑌𝑌_{𝑠𝑠𝑠𝑠𝑠𝑠𝑝𝑝𝑠𝑠𝑂𝑂𝑂𝑂𝑠𝑠𝑠𝑠𝑂𝑂𝑂𝑂} = 𝛽𝛽₀+ 𝛽𝛽_{𝑟𝑟𝑠𝑠𝑣𝑣𝑠𝑠𝑚𝑚𝑂𝑂}+ 𝜀𝜀 (Equation 4) 𝑌𝑌𝑠𝑠𝑠𝑠𝑠𝑠𝑝𝑝𝑠𝑠𝑂𝑂𝑂𝑂𝑠𝑠𝑠𝑠𝑂𝑂𝑂𝑂 = 𝛽𝛽0+ 𝛽𝛽𝑚𝑚𝑚𝑚𝑂𝑂𝑠𝑠𝑂𝑂𝑠𝑠 𝑂𝑂𝑂𝑂𝑖𝑖𝑂𝑂𝑥𝑥+ 𝛽𝛽𝑂𝑂𝑂𝑂𝑠𝑠𝑂𝑂𝑂𝑂𝑂𝑂𝑟𝑟𝑂𝑂𝑂𝑂+ 𝛽𝛽𝑟𝑟𝑠𝑠𝑣𝑣𝑠𝑠𝑚𝑚𝑂𝑂 + 𝛽𝛽𝐷𝐷𝐷𝐷𝐷𝐷𝑂𝑂𝑚𝑚𝑂𝑂𝑠𝑠+ 𝛽𝛽𝐷𝐷𝐷𝐷𝑝𝑝𝑠𝑠𝑚𝑚𝑝𝑝𝑚𝑚𝑂𝑂𝐷𝐷+ 𝛽𝛽_{𝐷𝐷𝐷𝐷𝑖𝑖𝑚𝑚𝐷𝐷𝑠𝑠} (Equation 5)

An overall measure of goodness of fit of the estimated regression line, denoted by 𝑅𝑅2_{, the}

coefficient of determination. If the slope (𝑅𝑅2_{) is equal to zero then it can be concluded that}

there is no relationship between the dependent variable Y and independent variables X (Gujarati, 2015).

The error term provides us with information that the model is not entirely accurate and implies that there are other factors than independent variables X that affects Y (Brooks, 2008)

Cover me, I&apos;m going public! : “The relationship between IPO and media coverage”