Terrorist attacks and their heterogenous effects on the financial sector in the Western Economies

(1)

1

Terrorist attacks and their heterogenous effects on the

financial sector in the Western Economies

By:

Lieke van Galen

Student number: 4141938

Student mail: l.van.galen@student.rug.nl

University of Groningen, Faculty of Economics and Business Pre-MSc. Finance

Supervisor: J. (Jos) Offerein, MSc Date: 29-5-2020

Abstract:

This paper analyses the impact of terrorist attacks on financial stock prices in Western Economies and the differences between these economies, using the event study

methodology. The impact of terrorist attacks is tested in the continents, Europe and America and is performed for the period 01-01-2000 to 31-12-2018. The study has shown that there is a negative effect within the financial sector on both the European- and the American stock market and that there is a difference between these Western Economies regarding terrorist attacks.

(2)

2

1. Introduction

September 11th _{2001, the day that a group of terrorists flew into the World Trade Center.}

The attack resulted in the closure of Wall Street, what led to a major decrease of the global stock market after it reopened. After 9/11, panic wandered around the world. The attack had a huge impact on the world population, about 3,000 people died because of this terrorist attack.

During the years after, the news of a new terrorist attack is travelling faster than ever, this causes that people are more aware of terrorism than ever before. Before the 9/11 attack the confidence of consumers from America had started to recover. However, after the attack this recover held back because of the psychological impact of the attack. What resulted in a negative effect on the growth in America. The uncertainty caused higher unemployment and lower expenditures and investments (Schlenger, 2004). Therefore, the effect of fear that a terrorist attack has on people, causes an effect on the whole economy (Brück & Wickström, 2004). A similar response occurred with the subway bombings in London on 7 July 2005. Both stock markets collapsed after the terrorist attacks, which resulted in significant financial consequences (Brounen & Derwall, 2010).

European and American financial structures, regardless of their differences, have been following similar trends in the 1990’s regarding that they both became more bank based. However, the difference in financial structure between the continents widened over time since 2000. The European financial structure consists of bank loans whereas the American financial structure consists of capital market instruments which include stocks, bonds and so on. During the last years differences have widened even more, Europe got a larger banking sector and America got a larger capital market (Cull & Pería, 2018). Seeing these changes over the previous years regarding the Europe bank loans and Americas capital market, it is expected to see a different reaction on the financial stock market looking at terrorist attacks. This event study will be the first that investigates this relation.

(3)

3

2. Literature overview

Central to the methodology that is used for this event study is the “efficient market

hypothesis”, the EHM was developed by Fama (1970). The main idea of the EMH is that an efficient capital market will have all the relevant available information directly fully reflected in the stock price of a certain company or companies. The EHM is the basis for an event study where abnormal returns are the result of the stock market reaction from the new information available in the market (McWilliams & Siegel, 1997). According to this theory, the stock market prices will change after a terrorist attack because of the new public information.

Many researchers investigated the impact of terrorist attacks on the stock market. There are many researchers that have concluded that terrorist attacks have a negative significant effect on stock markets, but that this negative effect lasts for a short time period (Arin, Ciferri & Spagnolo, 2008; Brounen & Derwall, 2010; Chen & Siems, 2003; Essaddam & Mnasri, 2015). Also, Chesney et al. (2011) concluded that the effects of a terrorist attack can be seen only right after the attack, so companies are not suffering from long lasting effects. Similarly, Aslam & Kang (2015) state that terrorist attacks affect the stock market, however the impact is limited. Other researchers have focused on the economic consequences and costs of terrorism. Gardeazabal & Abadie (2008) and Karolyi & Martell (2006) investigated the effect on the economy regarding terrorism. They both conclude that the impact of a terrorist attack is related to the location where the attack takes place. They are stating that the stock prices in wealthier countries are more affected after terrorist attacks, than in poor countries. Further they conclude that terrorist attacks result in negative consequences for the stock market prices, however they are relatively small. Carter & Simkens (2004) and Kolaric & Schiereck (2016) investigated the impact of 9/11 on airline stocks. They found negative significant abnormal returns for all studied airline company.

(4)

well-4 developed countries diminish more quickly. Following these studies, the stock returns are negatively affected by terrorist attacks in both Europe and America.

The European financial structure consists of bank loans. Due to the high bank loans, people respond disproportionately to macroeconomic events. This can be explained by the systemic risk being much higher in bank-based financial structures (Langfield & Pagano, 2016). Based on this information, the following hypothesis is composed:

H11: There is a significant, negative effect on the European stock market regarding

terrorist attacks.

As mentioned earlier, the stock returns will be negatively affected by terrorist attacks in the Western Economies. Financial markets are directly and indirectly victimised by terrorism. The financial markets in America consists of capital market instruments like stocks and bonds. For this reason, we also expect an effect on the American stock market. Therefore, the second hypothesis is:

H21: There is a significant, negative effect on the American stock market regarding

terrorist attacks.

Tengdong et all (2013) investigated the relationship between stock markets for different countries in different continents. They found that the investigated countries respond asymmetrically to shocks in the economy, however one faster than the other. Also Hilscher & Nosbusch (2010) investigated the effect of macroeconomic events, for example terrorist attacks, on countries and they found that the events have a statistically significant effect on the different stock markets. Due to larger fluctuations in countries terms of trade. The actual impact of a terrorist attack can differ between Europe and America, based on the differences in financial structures. That is why the main hypothesis of this paper is about the difference between the Western Economies. Therefore, the last hypothesis states that:

H31: There is a significant difference between the stock price reactions to a terrorist attack

(5)

5

3. Data and Methodology

3.1 Data

There are two data sources used for this paper. The information about the terrorist attacks is collected from the Global Terrorism Database. The GTD1_{is a database which provides}

information on international terrorist events around the world. The restrictions used for this paper are, that the attack occurred in Europe or America, between 2000 and 2018 (this restriction is chosen since the European index started at 31st_{December 1999) and with 10}

deaths or more (only deaths of the victims and the attackers, no injuries are included). Using these restrictions, the data will include 220 events. These 220 events are reduced to 102 (see appendix 2) by grouping the overlapping events, now we can calculate the variance of the abnormal returns without concern about the covariances between securities, because they are zero (Bernard, 1987). The terrorist attacks with 10 deaths or more are chosen because the small attacks are cancelled. A summary of the data is shown in the table below.

Table 1: Summary statistics

Summary Value

# terrorist attacks 102

# terrorist attacks in Europe 56

# terrorist attacks in America 46

Median deaths per terrorist attack 17

Average of deaths in Europe 32,4

Average of deaths in America 31,6

# of companies 18

# American companies 9

# European companies 9

The information about the stock market data is collected from Yahoo Finance. The information is collected from December 31st ₁₉₉₉2_{until December 31}st _2018.

1_{The GTD defines a terrorist attack as “the threatened or actual use of illegal force and violence by a non-state}

actor to attain a political, economic, religious, or social goal through fear, coercion, or intimidation”.

(6)

6 3.2 Methodology

The event study methodology, as discussed in the MacKinlay (1997) paper is used for this study. Terrorist attacks are unexpected events and can be used for the event study

methodology. It is unclear when a terrorist attack will happen, however when it happens the information about the event will be spread through the (social) media. This makes the identification for the attack easy. Therefore, terrorist attacks are a good measure for an event study.

The main idea of the event study methodology is to measure an abnormal return when an event happened. The abnormal returns are calculated using the market model (market- and risk-adjusted returns). The market model relates the return of a stock to the return of the market portfolio. First, there is an 80 days estimation window used to estimate the

parameters of the market model and these are used to calculate the normal returns in the event window. The event window has 41 days in total, which can be found in figure 1:

Estimation window Event window

-100 -20 0 20

Figure 1: Estimation – and event window

As first the estimated alpha and beta can be calculated as:

𝑅_𝑖𝑡 = 𝛼_𝑖 + 𝛽_𝑖 ∗ 𝑅_𝑚𝑡+ 𝜀_𝑖𝑡 (1) Where 𝑅𝑖,𝑡 is the stock return on day 𝑡 on event 𝑖 with 0 as the event date, 𝛼𝑖 is the constant

parameter, 𝛽𝑖 is the slope parameter, 𝑅𝑚𝑡 means the actual reference market portfolio

return at time 𝑡 and 𝜀𝑖𝑡 is a random variable (error term) with an expectation value of 0.

Then the abnormal returns can be calculated as:

𝐴𝑅_𝑖𝑡 = 𝑅_𝑖𝑡− 𝛼̂_𝑖 − 𝛽̂_𝑖 ∗ 𝑅_𝑚𝑡 (2) Where 𝐴𝑅𝑖𝑡 means the abnormal return of the stock day 𝑡 on event 𝑖 with 0 as the event

date, 𝑅𝑖𝑡 is the stock return on day 𝑡 on event 𝑖, 𝛼̂𝑖 is the OLS estimator of the constant

(7)

7 The reference market portfolio that is used for the American stock market is the S&P 500 index. This index, in which 500 large companies are included, gives a good interpretation of the American stock market. The individual stocks that are used are all included in the S&P500. The reference market portfolio that is used for the European stock market is the Euronext 100 index. The index is used because it gives a good interpretation of the European stock market, since the index includes the largest and most liquid stocks. The individual stocks that are used are included in the Euronext 100. Both individual company stocks (American – and European stock market) are only financial companies.

The next step is to calculate the average of all the abnormal return to test if the terrorist attacks have a negative effect on the American – and European stock market. The average abnormal return can be calculated as:

𝐴𝐴𝑅_𝑡= 1

102∑𝑖=1𝐴𝑅𝑖𝑡 (3)

To measure the total impact of a terrorist attack over a period (event window), we add up the individual abnormal returns, the cumulative abnormal returns (CAR), and is calculated as follows:

𝐶𝐴𝑅_𝑖(𝑡₁, 𝑡₂) = ∑𝑡2 𝐴𝑅_𝑖,𝑡

𝑡=𝑡1 (4)

Next the cumulative average abnormal returns (CAAR) can be calculated as follows: 𝐶𝐴𝐴𝑅(𝑡1, 𝑡2) =

1

102∑ 𝐶𝐴𝑅𝑖(𝑡1, 𝑡2) 𝑁

𝑖=1 (5)

3.2.1 Normal distribution test

To test the significance of the cumulative abnormal returns, we first need to check whether the cumulative abnormal returns have a normal distribution or not. The Jarcque-Bera test checks whether there is a normal distribution or not. The null hypothesis is that there is a normal distribution and the alternative hypothesis states that there is not a normal distribution. The Jarcque-Bera test can be calculated as follows:

𝐽𝐵 =102−𝑘+1

6 (𝑆 2 ₊1

4(𝐾 − 3)

2₎ ₍₆₎

(8)

8 When there is a normal distribution, the hypothesis can be tested using the T-test. When the returns do not have a normal distribution, the hypothesis can be tested through a non-parametric test, the Cowan (1992) or Kolari and Pynnönen (2010) test. In this case the CAAR’s are not normally distributed (see appendix 1) so, there is also a non-parametric test used in this event study to give more strength to the results. In addition to that, the

Generalized Sign Test (Cowan, 1992) is used for robustness. First, the Cross-Sectional T-test is calculated because the outcomes are related to the non-parametric test. The T-test is calculated as follows:

𝜃 = 𝐶𝐴𝐴𝑅(𝑡1,𝑡2)

𝑆𝑇𝐷 (𝐶𝐴𝐴𝑅(𝑡1,𝑡2)) (7) Where 𝑆𝑇𝐷 stands for the standard deviation of the CAAR, which is calculated as:

𝜎 = √ 1

102∑ (𝑥𝑖− 𝜇) 2 𝑛

𝑖=1 (8)

Where 𝑥𝑖 are the observed values of the sample and where 𝜇 is the mean of these

observations.

In addition to that, the Cowan Generalized Sign Test is used to test the significance of the European and the American stock market. The Cowan Generalized Sign Test examines the stocks with negative AR’-s in the event window compared to a period unaffected by the event. The negative AR’-s are given a score of 1 and the positive AR’s have a score of 0. The negative abnormal return proportion (parameter 𝑝̂) is calculated as follows:

𝑝̂ = 1

𝑛∑ 𝑆𝑖𝑡 𝑛

𝑖=1 (9)

Where 𝑛 is the number of abnormal returns and 𝑆𝑖𝑡 is the fraction of all negative abnormal

returns of the terrorist attacks. Next the Cowan Generalized Sign Test, test statistic is calculated as: 𝑧_𝐺 = 𝑤−𝑛𝑝̂ (𝑛𝑝̂(1−𝑝̂)) 1 2 (10)

(9)

9 Subsequently, to calculate whether the differences between the Western Economies are significant the non-parametric Kruskal-Wallis test is used. The Kruskal-Wallis test is calculated as follows: 𝑘 = [ 12 𝑛(𝑛+1)∑ 𝑇_𝑖2 𝑛𝑗 𝑘 𝑗=1 ] − 3(𝑛 + 1) (11)

Where 𝑘 is the test statistic of the Kruskal-Wallis test, 𝑛 is the total number of observations and 𝑇_𝑖2 means the square of the sum of the ranks.

To support the Kruskal-Wallis test, the Whitney-U test is also performed. The Mann-Whitney-U test can be calculated as follows:

𝑈 = 𝑛₁𝑛₂ +𝑛2(𝑛2+1)

2 ∑ 𝑅𝑖 𝑛2

𝑖=𝑛1+𝑥 (12)

Where 𝑈 is the Mann-Whitney-U test statistic, the 𝑛1𝑛2 is the total number of observations

(10)

10

4. Results

In this paper, the statistical tests are used to calculate the significance of the three

hypotheses. These statistical tests are calculated and stated in this section of the paper. The statistical tests are based on the calculated AAR’-s and CAAR’-s (table 2).

Table 2: Market Model AAR and CAAR America and Europe, largest event window

Market model Event -20 -19 -18 -17 -16 -15 -14 -13 -12 AAR Am -0,03% -0,24% -0,12% 0,12% 0,00% -0,07% -0,23% 0,05% -0,29% CAAR Am -0,03% -0,27% -0,39% -0,27% -0,27% -0,34% -0,57% -0,52% -0,80% AAR EU -0,30% -0,27% -0,31% 0,09% 0,01% -0,09% -0,29% -0,01% -0,12% CAAR EU -0,30% -0,57% -0,88% -0,79% -0,78% -0,87% -1,16% -1,17% -1,29% Event -11 -10 -9 -8 -7 -6 -5 -4 -3 AAR Am -0,24% -0,22% -0,09% -0,09% -0,16% 0,27% 0,19% -0,18% -0,09% CAAR Am -1,04% -1,27% -1,35% -1,44% -1,60% -1,33% -1,14% -1,32% -1,41% AAR EU -0,06% -0,36% -0,16% -0,24% -0,19% 0,13% 0,20% -0,19% 0,01% CAAR EU -1,35% -1,71% -1,87% -2,11% -2,30% -2,17% -1,98% -2,17% -2,16% Event -2 -1 0 1 2 3 4 5 6 AAR Am -0,02% 0,09% -0,10% -0,03% 0,14% -0,14% -0,03% -0,11% -0,14% CAAR Am -1,43% -1,33% -1,43% -1,46% -1,32% -1,46% -1,49% -1,60% -1,73% AAR EU -0,16% 0,16% -0,19% 0,00% -0,06% -0,24% -0,31% -0,15% -0,15% CAAR EU -2,33% -2,16% -2,35% -2,36% -2,41% -2,65% -2,96% -3,11% -3,26% Event 7 8 9 10 11 12 13 14 15 AAR Am -0,22% -0,20% -0,05% -0,43% -0,20% -0,03% 0,18% 0,12% -0,22% CAAR Am -1,95% -2,15% -2,20% -2,63% -2,83% -2,86% -2,69% -2,57% -2,78% AAR EU -0,24% -0,31% 0,02% -0,53% -0,20% 0,05% -0,08% -0,03% -0,21% CAAR EU -3,49% -3,81% -3,79% -4,32% -4,52% -4,48% -4,55% -4,58% -4,79% Event 16 17 18 19 20 AAR Am 0,14% -0,10% -0,09% -0,17% 0,00% CAAR Am -2,65% -2,74% -2,83% -3,00% -3,00% AAR EU -0,13% -0,24% -0,08% -0,09% -0,62% CAAR EU -4,92% -5,16% -5,23% -5,32% -5,94% 4.1 Main hypothesis

Firstly, the main analysis for the research question is mentioned and after that the other two hypotheses are analysed. The main analysis test includes the difference between the

(11)

11

Table 3: Kruskal-Wallis test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level

CAAR (-20, +20) CAAR (-10, +10) CAAR (-5, +5) CAAR (-2, +2)

America Europe America Europe America Europe America Europe

Rank Sum 2060 1343 610 293 143 110 37 18 Test Stat. 11,054 15,898 1,174 3,938 Critical Value: 1% 6,635 6,635 6,635 6,635 5% 3,841 3,841 3,841 3,841 10% 2,706 2,706 2,706 2,706

Significance YES*** YES*** NO YES**

Table 4: Mann-Whitney-U test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level

CAAR (-20, +20) CAAR (-10, +10) CAAR (-5, +5) CAAR (-2, +2)

America Europe America Europe America Europe America Europe

Rank Sum

482 1199 610 293 66 187 37 18

Test stat. -3,325 -3,987 -0,361 -1,984

P-value

0,001 6,686E-05 0,718 0,047

Significance YES*** YES*** NO YES**

(12)

12 hypothesis. There is a significant difference in the event window from -20, +20, -10, +10 and from -2, +2. These results correspond with the findings of other researchers about terrorist attacks. Both Tengdong et all (2013) and Hilscher & Nosbusch (2010) also reported that the countries investigated respond asymmetrically to macroeconomic events.

4.2 American stock market

Next, the outcomes of the tests about the effects on the American stock markets are mentioned. Table 5 and 6 show the outcomes of the T-test and the Cowan Generalized Sign test in terms of the CAAR (cumulative average abnormal returns) over different event date periods. For each period there are different results, they can be either statistically significant or not.

Table 5: Cross-Sectional T-test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level. (-20, +20) (-10, +10) (-5, +5) (-2, +2) CAAR/average -0,030 -0,016 -0,003 0,001 Standard Deviation 0,102 0,054 0,034 0,030 T-test Statistic -2,955 -2,989 -0,777 0,297

Significance YES*** YES*** NO NO

(13)

13

Table 6: Cowan Generalized Sign Test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level CAAR (-20, +20) (-10, +10) (-5, +5) (-2, +2) Proportion positive 0 0 3 2 Proportion negative 41 21 11 5 Test statistic 6,403 4,583 1,508 0,447

P-value 4,547E-13 4,768E-07 0,113 0,500

From table 5 and 6 we can see that the Cowan Generalized Sign Test and the T-test generalised the same results. The tests support each other, because there is a significant effect on the same CAAR event windows as the T-test.

4.3 European stock market

Next, the outcomes of the tests about the effects on the European stock markets are mentioned. Also, for the European stock market there is a T-test and a Cowan Generalized Sign Test conducted. For the same reason that the T-test is inconsistent with non-normal data distributions. Table 7 and 8 show the results for the third hypothesis.

Table 7: Cross-Sectional T-test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level. (-20, +20) (-10, +10) (-5, +5) (-2, +2) CAAR/ average -0,056 -0,030 -0,013 -0,003 Standard Deviation 0,106 0,073 0,038 0,031 T-test Statistic -5,396 -4,104 -3,444 -1,061

(14)

14

Table 8: Cowan Generalized Sign Test, where *** is significant at a 1% level, ** is significant at a 5% level and * is significant at a 10% level CAAR (-20, +20) (-10, +10) (-5, +5) (-2, +2) Proportion positive 0 0 4 1 Proportion negative 41 21 7 4 Test statistic 6,403 4,583 0,905 1,342

P-value 4,547E-13 4,768E-07 0,274 0,188

Also, from table 7 and 8 we can see that there are the same results from the Cowan

(15)

15

5. Conclusion

In this paper, we investigated the effect of terrorist attacks on the Western Economies, Europe and America, and the difference between these regions. The main hypothesis states that there is no significant difference between the stock price reactions in relation to a terrorist attack in Europe or America. The larger events windows have a significant result at the 1% level. Therefore, there is substantial evidence that the null hypothesis is rejected in these larger event windows. We can thus conclude that there is a significant difference between the stock price reactions to a terrorist attack in Europe and America. This means that the Western Economies react differently on a terrorist attack. This conclusion matches with the researchers Tengdong et all (2013) and Hilscher & Nosbusch (2010). This different reaction to terrorism between the continents can be referenced to the fact that America has a smaller banking sector and a larger capital market than Europe.

Furthermore, the first and second hypotheses are about the effects on the American- and European stock market regarding terrorist attacks. Both the T-test and the Cowan

(16)

16

6. References

Abadie, A., Gardeazabal, J. 2008. Terrorism and the world economy. European Economic

Review, 52(8): 1-27.

Aloui, C., Nguyen, D.K. 2014. On the detection of extreme movements and persistent behaviour in Mediterranean stock markets: a wavelet-based approach. Applied Economic letters, 46(22): 2611-2622.

Arin, K.P., Ciferri, D., Spagnolo, N. 2008. The price of terror: The effects of terrorism on stock market returns and volatility. Economics Letters, 101(3): 164-167.

Aslam, F., Kang, H.G. 2015. How different terrorist attacks affect stock markets. Applied

Economics letters, 26(6): 634-648.

Bernard, V.L. 1987. Cross-Sectional Dependence and Problems in Inference in Market-Based Accounting Research. Journal of Accounting Research, 25(1): 1-48.

Brounen, D., Derwall, J. 2010. The impact of terrorist attacks on international stock markets.

European Financial Management, 16(4): 585-598.

Brück, T., Wickström, B.A. 2004. The economic consequences of terror: guest editors’ introduction. European Journal of Political Economy, 20(1): 293-300.

Carter, D. A., Simkens, B.J. 2004. The market’s reaction to unexpected, catastrophic events: the case of airline stock returns and the September 11th attacks. The Quarterly Review of

Economics and Finance, 44(4): 539-558.

Chen, A.H., Siems, T.F. 2004. The effects of terrorism on global capital markets. European

Journal of Political Economy, 20(2): 349-366.

Chesney, M., Ganna Reshetar, G., Karaman, M. 2011. The impact of terrorism on financial markets: an empirical study. Journal of Banking and Finance, 35(2): 253-267.

Cowan, A.R. 1992. Non-parametric event study tests. Review of Quantitative Finance and

(17)

17 Cull, R., Pería, M.S.M. 2013. Bank ownership and lending patterns during the 2008–2009 financial crisis: Evidence from Latin America and Eastern Europe. Journal of Banking &

Finance, 37(12): 4861-4878.

Essaddam, N., Karagianis, J.M. 2014. Terrorism, country attributes, and the volatility of stock returns. Applied Economics letters, 22(3): 87-100.

Fama, E.F. 1970. Efficient Capital Markets: A Review of Theory and Empirical Work. The

Journal of Finance, 25(2): 383-417.

Hilscher, J., Nosbusch, Y. 2010. Determinants of Sovereign Risk: Macroeconomic Fundamentals and the Pricing of Sovereign Debt. Review of Finance, 14(2): 235-262. Johnston, R., Nedelescu, O. 2006. The impact of terrorism on financial markets. Journal of

Financial Crime, 13(1): 7-25.

Karolyi, G.A., Martell, R. 2006. Terrorism and the Stock Market. International Review of

Applied Financial Issues and Economics, 2(2): 285-314.

Kolari, J.W., Pynnönen, S. 2010. Event Study Testing with Cross-sectional Correlation of Abnormal Returns. The Review of Financial Studies, 23(11): 3996-4025.

Kolaric, S., Schiereck, D. 2016. Are stock markets efficient in the face of fear? Evidence from the terrorist attacks in Paris and Brussel. Finance Research Letters, 18(7): 306-310.

Kollias, C., Papadamou, S., Stagiannis, A. 2010. Terrorism and capital markets: The effects of the Madrid and London bomb attacks. International review of Economics and Finance, 20(4): 532-541.

Langfield, S., Pagano, M. 2006. Bank bias in Europe: effects on systemic risk and growth.

Economic Policy, 31(85): 51-106.

MacKinlay, A. C. 1997. Event studies in economics and finance. Journal of economic

literature, 35(1): 13-39.

(18)

18 Schlenger, W.E. 2004. Psychological Impact of the September 11, 2001 Terrorist Attacks: Summary of Empirical Findings in Adults. Journal of Aggression, Maltreatment & Trauma, 9(1-2): 97-108.

Tengdong, L., Hammoudeh, S., Thompson, M.A. 2013. A momentum threshold model of stock prices and country risk ratings: Evidence from BRICS countries. Journal of International

(19)

19

7. Appendix

Appendix 1: Jarque-Bera test American stock market:

CAAR (-20, +20) CAAR (-5, +5) CAAR (-2, +2) AAR

Kurtosis -1,014 1,630 -2,740 19,794 Skewness -0,053 0,724 0,388 0,471 Jarque-Bera test statistic 68,512 16,899 142,566 1202,448 H0 reject at 1% 1% 1% 1% Normal Data No No No No

European stock market:

CAAR (-20, +20) CAAR (-5, +5) CAAR (-2, +2) AAR

(20)

20 Appendix 2: List of terrorist attacks

Event Date Country Continent Kill

1 2-7-2000 Russia Eastern Europe 50

2 29-7-2000 Colombia South America 10

35 23-12-2004 Honduras Central America & Caribbean 28

(21)

21

45 17-3-2007 Mexico North America 10

47 26-1-2008 Guyana South America 11

48 18-2-2008 Guyana South America 12

51 9-4-2009 Peru South America 13

53 11-10-2009 Venezuela South America 10

54 5-11-2009 United States North America 13

61 11-4-2011 Belarus Eastern Europe 13

62 22-7-2011 Norway Western Europe 69

66 31-1-2013 Mexico North America 37

73 2-5-2014 Ukraine Eastern Europe 42

82 7-1-2015 France Western Europe 12

89 22-3-2016 Belgium Western Europe 35

(22)

22

91 14-7-2016 France Western Europe 87

95 22-5-2017 United Kingdom Western Europe 23

96 17-8-2017 Spain Western Europe 14

100 23-4-2018 Canada North America 10

101 13-5-2018 Venezuela South America 20