Olympic news and attitudes towards the Olympics: A compositional time-series analysis of how sentiment is affected by events

Transcription

1 Olympic news an attitues towars the Olympics: A compositional time-series analysis of how sentiment is affecte by events Professor Terence C. Mills School of Business an Economics Loughborough University Leicestershire UK LE11 3TU T.C.Mills@lboro.ac.uk Dr Peter Dawson School of Economics, University of East Anglia Norwich Research Park Norwich Norfolk UK NR4 7TJ. Peter.Dawson@uea.ac.uk Dr Paul Downwar School of Sport Eercise an Health Sciences Loughborough University Leicestershire UK LE11 3TU p.ownwar@lboro.ac.uk Abstract The Olympic Games are consiere to be the most prestigious multi-sport event in the worl. However, with growing costs associate with hosting such events against a backrop of questionable economic benefits, an yet elation that follows from sporting success, a number of stuies have starte to aress its intangible or softer impacts as a justification for the investment. It is well known that sentiment plays a part in the evolving economic valuation of companies through the stock market. What is less well known is how news affects the sentiment towars major public investments like the Olympics. In this paper we consier, from the contet of the preevent stage of the 30 th Olympia, the relationship between attitues towars the Olympics an Olympic-relate news; specifically the ba news associate with an increase in the cost of provision, an the goo news associate with Team GB s meal success in Using a unique ata set an an event-stuy approach that involves compositional time-series analysis, it is foun that goo news affects sentiments much more than ba but that the istribution of such sentiment varies wiely. For eample, a much more pronounce effect of goo news is ientifie for females than males, but ba news has less of an impact on the young an oler age groups. The paper consequently argues that etreme caution shoul be eercise in policy pronouncements that are base on such sentiments. Keywors: Attitues, news, Olympic Games, JEL coes: C25, D12, L83

2 Introuction The Olympics, an the Summer Olympics in particular, are a truly global event an are consiere to be the most prestigious multi-sport event in the worl. Competition to host the Olympics is fierce an, perhaps as a irect consequence, many prospective caniate cities make substantial claims about the economic, social an sporting impact of the Games. In the buil up to biing for, an hosting, the Lonon Olympics, UK sports policy aopte this approach. For eample, the Blair Government set out the following strategic priorities for sport in Game Plan : We conclue that government shoul set itself two overarching objectives: a major increase in participation in sport an physical activity, primarily because of the significant health benefits an to reuce the growing costs of inactivity; an a sustainable improvement in success in international competition, particularly in the sports which matter most to the public, primarily because of the feelgoo factor associate with winning. (DCMS/Strategy Unit,2002,12) These objectives were consiere to be mutually relate because of the limite possibility of a irect trickle-own effect from elite sports success to participation through the sports evelopment pyrami (Hinson et al., 1994; Hogan an Norton, 2000; Frawley an Cush, 2011). Consequently, the policy propose creating a twintrack approach to eveloping sports participation an elite sports success because of the evience of the health benefits that can be erive from the former, an espite the ifficulty in quantifying the impact, there appears to be a positive impact for the nation as a whole in the latter case (DCMS/Strategy Unit, 2002, 75). This paper aresses this latter claim by eamining the changes in the state monthly support, as measure as a scale variable, for hosting the 2012 games over 1

3 time. In particular the analysis focuses on the impacts of the ba news of cost overruns for 2012 announce in March 2007, an the goo news of Team GB s unepecte Beijing Olympic success in August The impacts of these news events are reveale using compositional time-series techniques that, unlike traitional time-series methos, account for the restrictions place on the ata by the use of a scale in eliciting opinion: in other wors, that increases in the proportion of one category of response must be matche by reuctions elsewhere on the scale. The results inicate that the collective subjective value of such events is contingent on the contet, so that, as a consequence, the policy an its unerpinning claims are potentially suspect. This matters because such subjective measures of the impact of public investment in sports are now more reaily cite as a justification for that investment, primarily because of the etremely weak evience in favour of their economic benefits. To illustrate these arguments, the net section briefly reviews the literature on the subjective impacts of major sports events. The ata an methoology employe in the research are then presente. This is then followe by the presentation of results an a iscussion of their implications. It is foun that goo news affects sentiment much more than ba but that the istribution of such sentiment varies wiely. For eample, a much more pronounce effect of goo news is ientifie for females than males, an ba news has a lesser impact on the younger an oler age groups. The paper consequently argues that etreme caution shoul be eercise in making policy pronouncements that are foune on ientifying positive sentiments from hosting or being successful at major sporting events. 2

4 Literature Review Despite the recent euphoria associate with Team GB s 2012 Olympic meal success, it remains that hosting events like the XXX Olympia require significant investments from the public sector, an ecisions to host them consequently shoul be base on a soun evience base. Significantly, the acaemic literature has generally been sceptical over the economic case that is typically mae for hosting such events. This case has traitionally reste on the claim that, through multiplier effects, positive economic impacts are erive, espite initial large-scale public subsiy. Typically such a case is mae by consultancy organisations, in avance of the Games, base on e ante forecasts. It has been shown that such evience is etremely questionable because of methoological flaws that populate the analyses that are, in turn, typically provie by business consultants working for lobby groups (Crompton, 1995, 2006). E post evaluation of the impacts has been unertaken by acaemics. These stuies apply an event-stuy framework to time-series ata, an eamine whether there are ajustments in the tren of key economic inicators after the major sports investment has taken place. The results suggest negligible an even negative impacts through crowing out (Baae, 2007; Downwar et al., 2009; Dawson, 2012). For eample, focusing on the 1984 Los Angeles an 1996 Atlanta Summer Olympics, Baae an Matheson (2002) foun that the (e-post) impact on unemployment in both regions was entirely transitory an conclue that long-run changes to steay-state equilibrium were only possible when new infrastructure an facilities were appropriate for the present an future economy. Maen (2006) an Giesecke an Maen (2007) conclue that initial e-ante assessments of the impact of the

5 Syney Olympics over-estimate the financial gain because the assumptions relating to inuce tourism an the responsiveness of the labour market were overly optimistic. It is partly because of such ifficulties that the claim that major sporting events such as the Olympics can provie important intangible benefits has emerge (Crompton, 2004). It is not surprising, therefore, that, as note in the introuction, the Lonon 2012 bi was base on the promise of using the Olympic Games to help eliver its twin-track strategy of promoting sports participation an generating national feelgoo, or subjective well-being. As this paper focuses on the latter issue, this begs the question of how hosting events can contribute towars success at them, an how that success affects the wier community. From the point of view of hosting events, it is well establishe that a home avantage eists (Balmer et al., 2001; Bernar an Busse, 2004; Shibli et al., 2012). It is likely that this follows from a combination of psychological factors that raise the prouctivity of home-base competitors, that home-base competitors have more familiarity with the facilities, an that the intensity of crow noise can influence officials through the pressures eerte upon them when making ecisions (for a review see Downwar et al., 2009). An increase in government spening an a policy priority on sport performance also precees the hosting of games (Houlihan an Green, 2008; Green an Houlihan, 2005). In the case of ientifying how the impact of hosting the Games an being successful at them impacts upon wier community sentiment, this is likely to be through spectatorship an be communicate by both attenance at the events an, for the vast majority of spectators, through meia eposure. The 29 th Olympia hel in 4

6 Beijing, for eample, witnesse an unpreceente amount of coverage, with a reporte 61,700 hours aire globally across 220 territories, generating a global auience of 3.6 billion (IOC, 2009). From the perspective of the UK, 41.1 million (approimately 67 per cent of the population) watche at least fifteen minutes of the BBC s Olympic broacast. In Lonon 2012 this increase to 51.9m (90% of the UK population) watching at least fifteen minutes of coverage (BBC, 2012). That sentiment has an effect on the public valuation of the Games is irectly aresse in a relatively small literature which, in part, motivates the current research. A number of stuies, for eample, have eamine the impact of the announcement of hosting the Games, or failing to be successful in biing to host them, on stock market valuations. The implication is that the perception that the games provie profitable business opportunities, as well as an increase in general optimism, can improve stock values. Such results have been ientifie in stuies of the Olympics by Veraros et al. (2004), Kavetsos an Szymanski (2008), Lees et al. (2009) an Downwar et al. (2010). That the impacts are not just meiate through a forecast of business opportunities is inicate by the fact that national team success can generate optimism an share price ajustment (Ashton et al., 2003). A further stran of literature has eamine the impacts of hosting an success at major sporting events on subjectively state well-being. Eamining a number of major events as well as the Olympics, Kavetsos an Szymanski (2010) ientify that hosting rather than success at sporting events increases the subjective well-being of citizens. Süssmuth et al. (2010) eamine the willingness of German citizens to pay to host the football Worl Cup in Willingness to pay is ientifie with the monetary value place on the civic prie associate with the event, espite non- 5

7 attenance at a venue. It is shown that eperience of hosting the football Worl Cup raises the willingness to pay from prior values an closes the gap between the valuations of former Western an Eastern German citizens. More common are stuies that have investigate the views of resients in the host region before an after (Ritchie an Aitken, 1985; Ritchie an Lyon,1990; Mihalik, 2003; Guala, 2009) or uring (Hiller an Wanner, 2011) the event. The attitues of responents beyon the host region, for eample aroun the time of the announcement (Atkinson et al., 2008; Walton et al., 2008) or uring the post-event perio (Frawley an Cush, 2011), have also been unertaken. Ritchie et al. (2009) provie one eample of a non-host perspective uring the preevent perio. They consier resient attitues in the communities of Weymouth an Portlan, locate in the county of Dorset an where the Lonon 2012 sailing events took place. They fin that, whilst resient attitues towars the impening event were in general supportive, there were significant negative effects ientifie, these being associate with likely congestion as well as concerns about potential increases in the cost of living. Though relatively unevelope as a literature, the above stuies suggest why UK sports policy might take a broaly positive view of hosting major events like the Lonon Games in terms of the feelgoo benefits that might be erive. However, it shoul be note that there is an asymmetric or positive bias enemic to the literature in that it tens to focus on success in either sporting terms at events or winning the bi to host an event. Significantly, once the research moves beyon that basic contet then the results o not appear so clear cut. This is implie in stuies of resients that suggest events might also impose costs. 6

8 It follows that the potential for news stories to affect public opinion cannot be unerestimate. Significant (negative) public opinion to both the Toronto 2008 an Chicago 2016 bis contribute towars the Games being hel elsewhere, whereas, in the buil up to the Athens 2004 Olympics, significant elays in the construction of venues an infrastructure attracte a lot of criticism. It has been argue that public opinion towars the value of hosting the Games tens to ip from just before winning the bi to the weeks leaing up to the Games (Cashman an Hughes, 1999). On the other han, it has also been ientifie that negative feelings towars the event can be transforme once the event takes place (Hiller an Wanner, 2011). In this stuy, therefore, it is hypothesise that the same sort of impact of news stories coul take place uring the preparation stage, an consequently the impact of the two most significant Olympic-relate stories that occurre uring the run-up to Lonon 2012 are eamine. These are news of the substantial increase in cost (March 2007) an the success of the British team uring the Beijing 2008 Olympics (August 2008). In respect of these issues, initial estimates by the Lonon Organising Committee place the total cost of hosting the event at 2.4 billion. However, in March 2007 it was announce that the total cost ha almost quaruple to just uner 9.3 billion (Government Olympic Eecutive Annual Report, 2011). In the 2008 Beijing Olympics, the British team won 47 meals, incluing 19 gol, finishing fourth in the meal table, which was a position that was wiely unepecte. Confirmation of better than epecte performance is provie by the UK Sporting Preferences Survey 2008 (UK Sport, 2009), who carrie out a pre-beijing survey asking interviewees where in the Olympic meal table they epecte the British team to finish. Only 16 per cent of responents epecte the team to finish in the top five. 7

9 Further verification is provie by a number of econometric moels of the etermination of meal counts that generally, an in some cases significantly, unerpreicte Britain s meal total (Forrest et al., 2010). Data an Methoology This stuy uses ata from the Taking Part Survey (TPS), commissione by the Department for Culture, Meia an Sport (DCMS). The TPS is a continuous (repeate cross-section, but not a panel) national survey of Englan which was first unertaken in In this stuy, ata from the first four waves (2005-6, , an ) are employe, generating over 35,000 usable observations. The TPS provies a unique opportunity to investigate attitues an behaviour uring the preparation stage of the Olympics, as these waves of the survey have a question specifically asking responents about their feelings towars the UK hosting the 2012 Summer Olympic Games in Lonon. Five options are presente in a measurement scale: strongly supportive ( 5 ), slightly supportive ( 4 ), neither supportive or against ( 3 ), slightly against ( 2 ), strongly against ( 1 ). t t t t t Importantly, the survey is aministere in such a way that there are enough observations to consier month-by-month variations in public attitues, rather than simple wave-specific (yearly) effects. This provies a rich ata set that allows a more precise investigation of the relationship between Olympic-relate news an public opinion. A preliminary view of the attitues an motivation of responents on a month-by-month basis (from January 2006 to March 2009) is presente in Figure I. 8

10 Figure I: Feelings Towars the Olympics Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Panel (a): Slightly or Strongly Against Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Panel (b): Slightly or Strongly Supportive 9

11 Panel (a) shows the proportion of responents against the UK hosting the 2012 Olympics an Panel (b) the proportion of responents who are supportive. Visual inspection suggests the presence of longer term trens though there appears to be a spike in the ata associate with cost overruns for those against hosting the games in Panel (a). Likewise in Panel (b) against a eclining tren, a spike in support for hosting the Games might be seen to follow from August To unpick these effects more forensically, a time-series analysis is unertaken, taking account of the compositional nature of the ata as implie by the measurement scale. Compositional Data A compositional ata set is one in which the n observations on 1 variables, enote,, 2,, i 1,2,, 1, are such that 0 an X i i1 i in 1, where 1 is the unit vector, here of imension 1. The matri then lies in the -part simple, i 1 i complete ata matri. i S, with. being the vector of fill-up values: X X 1 is thus the Because of the ifficulties of analysing X within the simple sample space, Aitchison (1982) propose mapping X from S to the -imensional real space eamining the statistical properties of the transforme ata within R an then R. Several transformations have been propose for oing this, the most popular being the aitive-logratio transformation efine as 10

12 Y y y y X 1 2 log log 1 2 a log (1) The inverse transformation, known as the aitive-logistic, is 1 ep y1 ep y 2 X a Y 1 ep y i 1 i 1 ep y i 1 i 1 ep y i 1 ep y i i 1 ep y i (2) Thus, an analysis which is ifficult in the awkwar sample space transforme using a to performe before using S may be R, in which more tractable statistical analyses may be 1 a to return to the original variables. Moelling Compositional Time Series Consier the case where the i, i1, i 2, X t to be the tth row of X an it i are time series of proportions, an are now written, so that there are T observations available. Accoringly, enote Y t to be the tth row of the aitive-logratio transforme ata matri Y. Brunson an Smith (1998) consier moelling Y t as a vector ARMA process but other frameworks are available in which covariates an tren functions may be introuce: see Mills (2007, 2009, 2010) for a variety of eamples. It shoul be emphasise that the moelling is invariant to the choice of fill-up value, so that any of the i can be chosen. 11

13 The survey ata available thus constitute a compositional time series of imension 1 5 of proportions from January 2006 to April 2009, a total of 40 observations, as efine earlier. They are plotte as Figure 2. Attention focuses on the two events of the March 2007 announcement of the increase in the cost of staging the Olympics an the success of Team GB at the Beijing Olympics of August These events are moelle as ummies, taking zeros everywhere ecept that value 1 in March 2007 an _ 08t takes the value 1 in August _ 07t takes the The proportions were aitive-logratio transforme using thus efining log y it it 5t 5 t as the fill-up value,, i 1,, 4. These ratios are shown in Figure 3 an were then moelle as a ummy-augmente vector AR(1) process, with this orer being selecte by a variety of criteria, the fitte moel being estimate as y y y y 1t 2t 3t 4t y 0.398y y y * , t 1 2, t 1 3, t 1 4, t 1 * u t _ 07 * t u t _ 08 * t u t * u t * * * (3) 12

14 Figure 2 Proportions having various levels of support for the Lonon Olympics (overall).6 Mar 07 Aug Jan 06 Jan 07 Jan 08 Jan Figure 3 Aitive log-ratios of proportions Jan 06 Jan 07 Jan 08 Jan 09 y1 y2 y3 y4 13

15 Stanar errors are in parentheses an coefficients significant at the 10% level are inicate by an asterisk. Few coefficients are iniviually significant an the moel is clearly over-parameterize, making interpretation problematic. The following restricte moel was therefore evelope. y y y y 1t 2t 3t 4t y y y y 1, t1 2, t1 3, t1 4, t u1t _ 07 t u t _ 08t u3t u t (4) The restrictions are either coefficient omissions or cross-equation coefficient equalities: in total a set of 19 restrictions are impose, with an accompanying likelihoo ratio statistic of 13.16, so the set is certainly acceptable. The nine coefficients that are freely estimate are all significant at less than the 1% level. The system is being riven by y, while both events appear to have ha 3 log 3 5 significant effects on the system, with y 1 an y 2 being positively affecte by the March 2007 event, an y 3 an y 4 being unaffecte, while all ratios are negatively affecte by the August 2008 event, movements that can clearly be seen in Figure 2. The system is more interpretable if log-proportions are consiere. The moel (4) can thus be written as 14

16 log 1t 0.96log3, t 1 log5t 0.04log5, t _ 07t _ 08t 1.42 log 2t 1.09log3, t1 log5t 0.09log5, t _07t 0.29 _08t 1.30 (5) log 3t 0.96 log3, t 1 log5t 0.04 log5, t _ 08 t log 4t 0.60log3, t 1 log5t 0.40log5, t _ 08t 0.20 Focusing on the impact of the two events, it can be seen that 1 an 2 were positively affecte by the March 2007 event, 3 an 4 were unaffecte an, by the nature of the composition, 5 was negatively affecte. In other wors, those proportions alreay hostile to the Olympics increase in response to the announcement that the games woul be more epensive, those proportions that were inifferent or milly supportive remaine the same, while the proportion strongly in favour ecline somewhat. Again these shifts can be seen in Figure 2. The response to the August 2008 event was that the first four proportions all ecline in size, in particular those who were inifferent, while the proportion most supportive increase in size, shifts that can also be seen in Figure 2. The performance of the British team in the Beijing Olympics thus ha a pronounce positive influence on support for the Lonon Olympics. The first row of Table 1 provies estimates of the changes in these proportions (measure in percentage points) obtaine via simulation of the moel. Thus the March 2007 event increase 1 an 2 by 2.3 an 2.4 percentage points respectively, with 5 being 15

17 corresponingly reuce by 4.7 points. The August 2008 event le to a ecline of 0.5, 0.6, 4.8 an 1.5 percentage points for the first four categories, with 5 corresponingly increasing by 7.4 points. Table 1 also presents analogous results at a isaggregate level by analysing the geographic groups ecluing Lonon (E Ln), north, south an milans an east (M&E), the age groups 16 29, 30 44, an 60+, an males an females. For brevity, etails of the moels are omitte. Table 1 Change in support in response to the two events (percentage points). March 2007 August All E Ln North South M&E Male Female

18 The results show some consierable variation in impact. Broaly speaking, the goo news of Olympic success prouces relatively large increases in strong support for the Olympics, as the above literature review might suggest. However, this is not uniform. It is most prevalent in the younger age-groups an females an more so away from Lonon. This might be because the Olympics reaches an auience that is usual for sport, that of young males, but it also attracts female support because, unlike most professional sports, it is not male ominate. That Lonon an the south seem to respon less to success coul be inicative of the concerns of resients of the host areas note earlier. Interestingly, the impact of the cost overruns being announce in March 2007 are accompanie by increases in support for the Games in the north an for females, which might suggest that, for the reasons note above, the benefits of hosting the games ecee the costs for these groups. In general, however, for males an those of mile age there is a clear shift away from support for the Games in the contet of ba news. Conclusions It is now increasingly common in public policy iscussion to make a case for investment in major sporting events like the Olympic Games that raws upon the softer an more intangible effects that follow from sporting success at, as well as hosting, such events. In particular it is argue that they both contribute to the feelgoo factor for society, that is, subjective well-being. The literature that supports such claims, however, tens to have an implicit bias in it as it focuses on one imension of the potential impact, that of success in either seeking to host an event, or through sporting success at the event. 17

19 Although it is well known that ifferences in sentiment play a part in the evolving economic valuation of companies through the stock market, it is less well known how news affects sentiment towars major public investments like the Olympics. In this paper the relationship between attitues towars the Olympics, the ba news associate with an increase in the cost of provision an the goo news associate with Team GB s meal success in 2008 are analyse. Using a unique ata set an an event-stuy approach that involves compositional time-series analysis enables us to ientify that goo news affects sentiment much more than ba news, but that the istribution of such sentiment varies wiely. The paper consequently argues that etreme caution shoul be eercise when making policy pronouncements that are base on such sentiment. 18

20 References Ashton, J.K., Gerrar, B. an Huson, R. (2003) Economic Impact of national Sporting Success: Evience from the Lonon Stock Echange, Applie Economics Letters, 10: Aitchison, J. (1982) The statistical analysis of compositional ata (with iscussion). Journal of the Royal Statistical society, Series B, 44: Atkinson, G., Mourato, S., Szymanski, S. an Ozemiroglu, E. (2008) Are we willing to pay enough to back the bi?: Valuing the intangible impacts of Lonon s bi to host the 2012 Summer Olympic Games. Urban Stuies 45: Baae, R.A. (2007) The economic impact of mega-sporting events. In: Anreff W an Szymanski S (es.) Hanbook of the Economics of Sport, Ewar Elgar: Cheltenham. Baae, R.A.,an Matheson V. (2002) Biing for the Olympics: fool's gol? In: Barros C P, Ibrahimo M, an Szymanski S (es.) Transatlantic Sport: the comparative economics of North America an European Sports, Ewar Elgar: Lonon. Balmer, N.J., Nevill, A.M. an Williams, A.M. (2001) Home avantage in the Winter Olympics ( ). Journal of Sport Sciences 19: BBC (2012) accesse 1 st October Bernar, A. an Busse, M. (2004). Who wins the Olympics? Economic resources an meal totals. Review of Economics an Statistics 86: Brunson, T.M. an Smith, T.M.F. (1998) Time series analysis of compositional ata. Journal of Official Statistics, 14:

21 Cashman, R. an Hughes, A. (1999). Staging the Olympics: The Event an its Impact.UNSW Press: Syney. Crompton, J. (1995) Economic Impact Analysis of Sports facilities an Events: Eleven sources of misapplication, Journal of Sports management, 9: Crompton, J. (2004) Beyon economic impact: an alternative rationale for the subsiy of major league sports facilities, Journal of Sports Management, 18: Crompton, J. (2006) Economic Impact Stuies: Instruments for Political Shenanigans? Journal of Travel Research, 45: Dawson, P. (2012) Economics of the Olympics. In Kahane, L an Shmanske, S (es.) The Ofor Hanbook of Sports Economics Volume 1: The Economics of Sports. Ofor University Press: New York. DCMS/Strategy Unit (2002) Game Plan: A strategy for elivering government s sport an physical activity objectives. Downwar, P.M. Dejonghe, T. an Dawson, A. (2009) Sports Economics: Theory, Evience an Policy, Elsevier: Lonon. Downwar, P. Nishio, T. an Lim, C. (2010) Olympics Host City Selection an its impact on Stock markets, Journal of Hospitality an Tourism, 1, Forrest, D., Sanz, I., an Tena, J.D. (2010) Forecasting national team meal totals at the Summer Olympic Games. International Journal of Forecasting 26: Frawley, S. an Cush, A. (2011) Major sports events an participation legacy: the case of the 2003 Rugby Worl Cup. Managing Leisure 16:

22 Giesecke, J., an Maen, J. (2007) The Syney Olympics, Seven years on: an e- post ynamic CGE assessment. Centre of Policy Stuies Monash University Working Paper G-168. Green, M. an Houlihan, B. (2005) Elite Sport Development: Policy Learning an Political Priorities. Lonon: Routlege. Guala, C. (2009) To bi or not to bi: Public opinion before an after the Games: The case of Turin. In Kennell, J., Blaon, C., an Booth, E. (es.) The Olympic Legacy: People, place, enterprise. University of Greenwich, Lonon. Hiller, H., an Wanner, R. (2011) Public opinion in host Olympic cities: The case of the 2010 Vancouver Winter Games. Sociology 45: Hinson, A., Gilow, B., an Peebles, C.(1994) The trickle-own effect of top-level sport: myth or reality? A cast stuy of the Olympics. Australian Journal of Leisure an Recreation 4: Hogan, K. an Norton, K. (2000) The price of Olympic gol. Journal of Science an Meicine in Sport 3: Houlihan, B. an Green, M. (es). (2008) Comparative Elite Sport Development: Systems, Structures an Public Policy. Ofor:Butterworth-Heinemann. IOC (2009) Games of the XXIX Olympia, Beijing 2008: Global Television an the Online Meia Report. Accesse from: Global_Broacast_Overview.pf [Accesse May 2012] Kavetsos, G. an Szymanski, S. (2008) Olympic Games, terrorism an their impact on the Lonon an Paris stock echanges. Revue Economie Politique 118:

23 Kavetsos, G. an Szymanski, S. (2010) National well-being an international sports events. Journal of Economic Psychology, 31: Lees, M., Mirikitani, J. an Tang, D. (2009) Rational euberance? An event analysis of the 2008 Olympics announcement. International Journal of Sport Finance 4: Maen, J. (2006) Economic an fiscal impacts of mega sporting events: a general equilibrium assessment. Public Finance an Management 6: Mills, T.C (2007) Moelling compositional ata in economics. Inian Economic Journal 55: Mills, T.C (2009) Forecasting obesity trens in Englan. Journal of the Royal Statistical Society, Series A, 172: Mills, T.C. (2010) Forecasting compositional time series. Quality an Quantity 44: Mihalik, B. (2003) Host population perceptions towars the 1996 Atlanta Olympics: Benefits an liabilities. In De Moragas, M., Kennett, C., an Puig, N. (es.) The legacy of the Olympic Games International Olympic Committee, Lausanne. Ritchie, J.R.B. an Aitken, C.E. (1985) OLYMPULSE II evolving resient attitues towars the 1988 Olympic Winter Games. Journal of Travel Research 23: Ritchie, J.R.B. an Lyons, M.M. (1990) OLYMPULSE IV A post-event assessment of resient reaction to the XVth Olympic Winter Games. Journal of Travel Research 28:

24 Ritchie, B. W., Shipway, R. an Cleeve, B. (2009) Resient perceptions of mega- sporting events: a non-host city perspective of the 2012 Lonon Olympic Games. Journal of Sport an Tourism 14, Shibli, S., Gratton, C. an Bingham, J. (2012) A forecast of the performance of Great Britain an Northern Irelan in the Lonon 2012 Olympic Games. Managing Leisure 17: Süssmuth, B., Heyne, M., an Maennig, W. (2010) Inuce civic prie an integration, Ofor Bulletin of Economics an Statistics, 72, UK Sport (2009) UK Sporting Preferences Survey Veraros, N., Kasimati, E., an Dawson, P. (2004) The 2004 Olympic Games announcement an its effect on the Athens an Milan stock echanges. Applie Economics Letters 11: Walton, H., Longo, A., an Dawson, P. (2008) Contingent valuation of 2012 Lonon Olympics: A regional perspective. Journal of Sports Economics 9: