The Relationship Between World Cup Events And Stock Market: A Critical Evaluation

The issues you may discuss include (but not limited to)
• Is this research question important and why?
• Is the research design sensible? If not, what can be done to improve?
• Is the basic data analysis done adequately? If not, what can be done?
• Is the econometric specification appropriate? If not, how can it be improved?
• Is the econometric methodology used adequate? If not, how can you improve it?
• Do the authors conduct adequate sensitivity analysis?
• Do the authors’ conclusions make economic sense and matters economically?

There are many factors influencing the stock market, these include events as unimagined as sports and even obvious ones such as war. Generally, there is a relationship between change in stock prices and investor sentiment. This is as demonstrated by Guy Kaplanski and Levi in their paper on “Exploitable Predictable Irrationality” taking into account the FIFA world cup event.( Kaplanski & Levi. 2010. Pp 535-553)

In the paper, the writers argue that losing and winning often shift the investor sentiment, which in the end affect their investment decisions. Naturally, there is only a single winner out of every football match encounter and therefore after every game the stock market gets affected.

To investigate this hypothesis, an ideal neutral market is chosen, in this case the US stock market. This is due to its versatility of global investors and therefore all the games played in the world cup are likely to affect it. In their findings, a conclusion that losing affects the stock market negatively whereas the effect of winning is negligible. 

One of the main hypotheses is that losing instigates a negative effect by the fans whose countries lose. To test this, three methods are employed, i.e.:

1. Theoretical, independent of match results
2. Comparing average returns on the stock market, given that during the game period. On about 30 teams lose.

• Technical analysis, through use of exact data to determine relationship in stock market and sporting periods

In brief, the paper focuses on sentimental influence on the investment decisions made by investors. Not only on local stock markets but also on international markets.

The major points drawn include:

1. US market has large number of investors compared to other markets
2. Sentiments affect investor decisions, hence stock market. Hence there is a relationship between stock market and investor sentiment

• Losing of country teams affects the stock market rather adversely

1. Winning has a negligible positive effect

Several scholars have carried out research in order to answer the question of whether there is a relationship between investor sentiment and investments. There have been a number of researches conducted such as:

1. Investor sentiment and stock returns (Fisher & Meir. 2000)
2. Sports sentiment and stock market returns (Sevil & Polat. 2015)

All these articles and many others seek to answer the same question as that rose in the research. It is therefore important to determine the factors that influence the investor decision since it is through this that the market may be modeled for prediction.  Interestingly, games are among instigators for sentiments, this is due to the loyalty attached to a particular team and even more the spirit of nationalism. Therefore, the research question is important in contributing to the learning of the mindset of the investor in relation to stock returns.

The research process is designed into six sections, each covering a specific aspect to consolidate the whole research process, i.e.:

1. Introduction
2. Background to the research motivation

• Methodology used in the research process

1. Results and interpretations
2. Recommendations based on the results
3. Conclusion

The study design is a standard research procedure; therefore, it captures most of the research aspects. However, inclusion of the areas into which the research features would improve the relevancy to the readers. I.e. explanation of the study context, hence it is sensible

Critical Evaluation

The method on data analysis involves regression to determine the relationship presented among the data variables; which include:

• Date of investigated events
• Return on stock
• Influential variables i.e. dummy variables

 The research involves multiple regressions i.e. R_i= Y_o+ βX_i + £_I  where R is the daily stock returns, Y correlation coefficient intercept and X_i are the independent variables robustness check and visual graphing, which is adequate to investigate multiple relations between interest variables and in carrying out data interpretation. This enables to establish the significant factors. The process of data analysis and output processing was done using Eview software.

Interestingly, despite the indirect relation between football and stock in America, the research pitches the economic specification to determine stock trends during world cup events. In the American stock market, this is because, according to the authors the fact that approximately a third of the companies quoted in the stock market are foreign and therefore the stock market is most likely to be influenced by foreign investor decisions. Therefore

In application of mathematics and statistics to economics, the authors set up an economic model used to show the effect of football sport during world cup events in a period after the Second World War and 2006. The model specification includes outliers so as to be realistic and in line with real life situation.

Model:

R_i= y_{o + 1i}R_i-1+ _2iD_ii+y₃H_i + y₄T_i + y₅P_i+ y₆E_i + _7iJ_ii+ ?_i

Where ( Rt =daily return,Y0= regression intercept coefficient, Rt-1, Rt-2 =

1^st and 2^nd preceding day returns, Dit , [1,2,3,4], are dummy days of the week: Monday, Tuesday, Wednesday, and Thursday,10 Ht is a dummy variable for days after a non-weekend holiday and Tt is a dummy for first five days of taxation year, Pt is a dummy variable for the world cup (June–July) event.)  (Kaplanski & Levi, 2010. Pp 545). 

The authors carry out a sensitivity analysis through:

1. Explaining the association between the inputted and outputted variables
2. Simplifying the regression model to remove insignificant structure parts

• Lowering uncertainty by identifying the inputs

1. Locating errors in the model
2. Testing model robustness
3. Investigating key connections in the observations and forecasts

Therefore, the researchers carry out adequate sensitivity analysis.

 After testing the null hypothesis, carrying out analysis and interpretation of the output, the authors find a relationship between stock prices during the world cup events and the winning or losing of teams. Therefore, they conclude that the world cup effect is:

• Large
• Highly significant
• Long-lasting

Such that the average return during the event is -2.58%, compared to normal days on the same period when returns average on +1.21%.

This conclusion is sensible and is important in determining when to buy and sell stocks as major sports events happen, it also suggests balancing on stocks bought to avoid bad losses during world cup event. 

Replication

In analyzing the relationship between stock returns from value-weighted index and from equal weighted index, we use market data, having variables:

i. Dates ii. Stock returns( RVW- value weighted index, REW- equal weighted index) iii. Dummy days- ((D1,D2,D3,D4,D5-mon, Tue, wed, Thurs, Fri), H-dummy variable for days after non-weekend holiday, T-dummy days for 1st  5 taxation year days, E-event days, J-dummy for 10 days with highest returns(1= highest, 2=lowest), P-control variable

 The analysis involves ordinary least squares regression and ANOVA in R. Output is then generated as:

##
## Call:
## lm(formula = RVW ~ E + D1 + D2 + D3 + D4 + D5 + P + T + J1 +
##     J2 + H, data = dat)
##
## Coefficients:
## (Intercept)            E           D1           D2           D3  
##   8.635e-04   -1.508e-03   -1.712e-03   -5.991e-04   -3.793e-05  
##          D4           D5            P            T           J1  
##  -3.201e-04   -3.835e-05    1.414e-05    2.335e-04   -7.719e-02  
##          J2            H  
##   6.634e-02    7.804e-04

## Analysis of Variance Table
##
## Response: RVW
##              Df  Sum Sq  Mean Sq  F value  Pr(>F)    
## E             1 0.00078 0.000783   9.9181 0.00164 **
## D1            1 0.00603 0.006034  76.3924 < 2e-16 ***
## D2            1 0.00042 0.000419   5.3048 0.02128 *  
## D3            1 0.00006 0.000061   0.7744 0.37888    
## D4            1 0.00017 0.000175   2.2154 0.13666    
## D5            1 0.00000 0.000000   0.0033 0.95426    
## P             1 0.00000 0.000000   0.0024 0.96086    
## T             1 0.00004 0.000036   0.4554 0.49977    
## J1            1 0.05964 0.059638 755.0297 < 2e-16 ***
## J2            1 0.04394 0.043944 556.3389 < 2e-16 ***
## H             1 0.00026 0.000256   3.2358 0.07206 .  
## Residuals 16807 1.32754 0.000079                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##                     2.5 %        97.5 %
## (Intercept) -1.004847e-03  0.0027317483
## E           -2.502098e-03 -0.0005129705
## D1          -3.606291e-03  0.0001821602
## D2          -2.491652e-03  0.0012933780
## D3          -1.930377e-03  0.0018545201
## D4          -2.213003e-03  0.0015727626
## D5          -1.931203e-03  0.0018545055
## P           -3.661584e-04  0.0003944327
## T           -7.349165e-04  0.0012018505
## J1          -8.270597e-02 -0.0716831654
## J2           6.082491e-02  0.0718473372
## H           -6.996404e-05  0.0016306675

##
## Call:
## lm(formula = RVW ~ E + D1 + D2 + D3 + D4 + D5 + P + T + J1 +
##     J2 + H, data = dat)
##
## Residuals:
##       Min        1Q    Median        3Q       Max
## -0.093379 -0.004051  0.000258  0.004377  0.080933
##
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  8.634e-04  9.532e-04   0.906  0.36501    
## E           -1.508e-03  5.074e-04  -2.971  0.00297 **
## D1          -1.712e-03  9.664e-04  -1.772  0.07648 .  
## D2          -5.991e-04  9.655e-04  -0.621  0.53491    
## D3          -3.793e-05  9.655e-04  -0.039  0.96866    
## D4          -3.201e-04  9.657e-04  -0.331  0.74028    
## D5          -3.835e-05  9.657e-04  -0.040  0.96832    
## P            1.414e-05  1.940e-04   0.073  0.94191    
## T            2.335e-04  4.941e-04   0.473  0.63653    
## J1          -7.719e-02  2.812e-03 -27.454  < 2e-16 ***
## J2           6.634e-02  2.812e-03  23.593  < 2e-16 ***
## H            7.803e-04  4.338e-04   1.799  0.07206 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.008887 on 16807 degrees of freedom
## Multiple R-squared:  0.07738,    Adjusted R-squared:  0.07678

Ordinary Least Squares

## F-statistic: 128.2 on 11 and 16807 DF,  p-value: < 2.2e-16

##
## Call:
## lm(formula = REW ~ E + D1 + D2 + D3 + D4 + D5 + P + T + J1 +
##     J2 + H, data = dat)
##
## Coefficients:
## (Intercept)            E           D1           D2           D3  
##   0.0014420   -0.0015814   -0.0026029   -0.0015233   -0.0003506  
##          D4           D5            P            T           J1  
##  -0.0003392    0.0004399   -0.0001446    0.0038606   -0.0742936  
##          J2            H  
##   0.0642939    0.0011366

## Analysis of Variance Table
##
## Response: REW
##              Df  Sum Sq  Mean Sq  F value    Pr(>F)    
## E             1 0.00113 0.001132  19.8745 8.323e-06 ***
## D1            1 0.01289 0.012892 226.4225 < 2.2e-16 ***
## D2            1 0.00472 0.004717  82.8424 < 2.2e-16 ***
## D3            1 0.00032 0.000323   5.6813  0.017158 *  
## D4            1 0.00113 0.001129  19.8209 8.560e-06 ***
## D5            1 0.00001 0.000010   0.1750  0.675733    
## P             1 0.00012 0.000121   2.1253  0.144903    
## T             1 0.00523 0.005230  91.8443 < 2.2e-16 ***
## J1            1 0.05525 0.055248 970.3085 < 2.2e-16 ***
## J2            1 0.04127 0.041269 724.7842 < 2.2e-16 ***
## H             1 0.00054 0.000542   9.5230  0.002032 **
## Residuals 16807 0.95698 0.000057                       
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##                     2.5 %        97.5 %
## (Intercept) -0.0001442574  3.028253e-03
## E           -0.0024258467 -7.370025e-04
## D1          -0.0042112063 -9.946690e-04
## D2          -0.0031301073  8.352524e-05
## D3          -0.0019573948  1.256125e-03
## D4          -0.0019462919  1.267965e-03
## D5          -0.0011671802  2.047028e-03
## P           -0.0004674439  1.783265e-04
## T            0.0030383836  4.682772e-03
## J1          -0.0789730126 -6.961424e-02
## J2           0.0596146273  6.897309e-02
## H            0.0004146666  1.858567e-03

##
## Call:
## lm(formula = REW ~ E + D1 + D2 + D3 + D4 + D5 + P + T + J1 +
##     J2 + H, data = dat)
##
## Residuals:
##       Min        1Q    Median        3Q       Max
## -0.062057 -0.003094  0.000523  0.003706  0.048395
##
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.0014420  0.0008093   1.782 0.074792 .  
## E           -0.0015814  0.0004308  -3.671 0.000242 ***
## D1          -0.0026029  0.0008205  -3.172 0.001515 **
## D2          -0.0015233  0.0008198  -1.858 0.063156 .  
## D3          -0.0003506  0.0008197  -0.428 0.668844    
## D4          -0.0003392  0.0008199  -0.414 0.679132    
## D5           0.0004399  0.0008199   0.537 0.591583    
## P           -0.0001446  0.0001647  -0.878 0.380197    
## T            0.0038606  0.0004195   9.204  < 2e-16 ***
## J1          -0.0742936  0.0023873 -31.120  < 2e-16 ***
## J2           0.0642939  0.0023872  26.932  < 2e-16 ***
## H            0.0011366  0.0003683   3.086 0.002032 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.007546 on 16807 degrees of freedom
## Multiple R-squared:  0.1136, Adjusted R-squared:  0.113
## F-statistic: 195.8 on 11 and 16807 DF,  p-value: < 2.2e-16

The conclusions drawn from the analysis include:

1. The world cup coefficient is positive for returns on stock from equal weighted index and for value weighted index
2. REW is the most affected during the world cup event days more than RVW

• The world cup coefficient is very large and enormously significant( t-value ranges from -6.371 to -31.120)

1. There is a negligible robust significance in most of  the test variables
2. There are few outliers in the models used, i.e. all the variables affected the equal and valued stock returns during the world cup event  

Economic theorists often argue that major multiregional boost local economies of the host countries, this includes:

1. International games such as world cup and Olympics
2. Trade conventions

• Democratic national conventions

This is suggested to be due to a gush in the economic activities connected to the event. (Ross. 2018) . Previous research shows that when a country wins an Olympic medal, national stock activity decreases (Jessica & Markellos, 2018).

The suggested reasons infer that games distract both the public and investors alike. The main hypothesis is that Olympic games affect the stock market of the host country and the international markets of other countries participating in the event. Our null hypothesis is that Olympic Games do not affect the stock market and that stock markets are independent of investor swayed sentiments.

The alternative hypothesis is that Olympic Games distract investors, where with every gold medal won, sales volume drop otherwise there is no effect in the sales. (Jessica & Markellos, 2018).

However, it is a topic of argument as to whether the Olympic Games affect attention or sentiments of the investors.

The dataset used is based on 4 summer Olympic Games):

1. Sydney-2000
2. Athens-2004

• Beijing-2008

1. London-2012

Six of the eight countries that have hosted the Olympics since 1984 saw their stock market generate positive returns. (Kiran, 2016

With variables:

1. Winsact- effect of winning a medal on stock activity
2. Normsact- effect of no medal won on stock market activities

• Watch- variables for watching game

1. No-watch- variable for not watching game
2. Effect- effect of watching the game

The output from analysis in r is:

##     Winsact**              Normsact**              Watch        
##  Min.   :-0.0529220   Min.   :-0.0530540   Min.   :0.00000  
##  1st Qu.:-0.0021010   1st Qu.:-0.0019510   1st Qu.:0.00000  
##  Median : 0.0009650   Median : 0.0009720   Median :0.00000  
##  Mean   : 0.0005748   Mean   : 0.0005946   Mean   :0.02132  
##  3rd Qu.: 0.0040170   3rd Qu.: 0.0037560   3rd Qu.:0.00000  
##  Max.   : 0.0225590   Max.   : 0.0212990   Max.   :1.00000  
##     No.watch          effect      
##  Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.0000   1st Qu.:0.0000  
##  Median :0.0000   Median :0.0000  
##  Mean   :0.1839   Mean   :0.1885  
##  3rd Qu.:0.0000   3rd Qu.:0.0000  
##  Max.   :1.0000   Max.   :1.0000

##
## Call:
## lm(formula = Winsact ~ Watch + No.watch + effect, data = olym)
##
## Residuals:
##       Min        1Q    Median        3Q       Max
## -0.049761 -0.002746  0.000350  0.003254  0.024852
##
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.0013787  0.0001922   7.175 1.13e-12 ***
## Watch       -0.0020595  0.0010468  -1.968   0.0493 *  
## No.watch    -0.0024803  0.0004010  -6.186 7.95e-10 ***
## effect      -0.0016118  0.0003971  -4.059 5.19e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.005858 on 1497 degrees of freedom
## Multiple R-squared:  0.03203,    Adjusted R-squared:  0.03009
## F-statistic: 16.51 on 3 and 1497 DF,  p-value: 1.471e-10

## Analysis of Variance Table
##
## Response: Winsact
##             Df   Sum Sq    Mean Sq F value    Pr(>F)    
## Watch        1 0.000134 0.00013381  3.8992   0.04849 *  
## No.watch     1 0.001001 0.00100090 29.1668 7.714e-08 ***
## effect       1 0.000565 0.00056534 16.4744 5.186e-05 ***
## Residuals 1497 0.051371 0.00003432                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

**Source: based on the Oxford Olympics data of  data.world, https://data,world/sports/olympics 

From the output, we can conclude that:

• Watching the game has effect on the distraction of the populace and hence the market sales activities are disrupted
• Winning a medal affects the subsequent day sales volume
• There is a significant relationship effect when watching the game, the country wins a medal and the sales volume

This supports the survey done in London (2012) games that indicate; one out of four people listen or watch the event while at work (University of East Anglia, 2018).

Therefore, from the research analysis we conclude, major sports events affect the investor attention therefore affecting the stock volume sales. 

Kaplanski, G., & Haim, L. (2008). Exploitable Predictable Irrationality: The FIFA World Cup Effect on the U.S. Stock Market. Journal of Financial and Quantitative Analysis, forthcoming ( 2008), 535-553.

Kenneth, F, L., & Meir S. (2000). Investor Sentiment and Stock Returns. Financial Analysts      Journal (vol. 30), No. 4:50-51. Retrieved from: https://www.cfapubs.org/doi/pdf/10.2469/faj.v56.n2.2340 

Sevil, T., Polat, A.(2015). Sport Sentiment and Stock Market Returns: An Evidence of National Match Days In Turkey. International journal of Economics, Commerce, and Management.(vol.3) issue 12. ISSN 23480386.

Kabtta, S, K.(2016). Stock Markets of Host Countries surge During Olympics. Retrieved                                    from: https://m.economictimes.com/markets/stocks/news/stock-markets-of-host-countries-surge-duringolympics/articleshow/53565277.cms, on 4^th April 2018.

Jessica Y. Wang et al.(2018) Is There An Olympic Gold Medal Rush In The Stock Market?, The European Journal of finance(2018). DOI: 10.1080/1351847X.2017.1421245.

University of Anglia. (2018, January 16^th). Study Shows How Olympic Games Affect The Stock             Market. Retrieved from: https://www.uea.ac.uk/about/-/study-shows-how-olympic-games-affect-the-stock-market