Xác suất thống kê: Phân tích số lượng tử vong bởi Covid-19 ở European Union và Europe [English version]

To sum up, it is obvious that the rate of total deaths in European Union is approximately equal to the value of total deaths in Europe.. o Average Rainfall X1 H0: b1 = 0 total death and

Business Statistics – ECON1193 Assignment 3 – Team Assignment Report - 3A

Table of Content:

Part 1: Data Collection (40 countries) on raw data Excel file

PART 2: Descriptive Statistics

Measurement of Central Tendency

Key findings:

 The mean of total deaths in European Union (0.00020) is smaller than this value

of total deaths in Europe ( 0.00027)

 The median of total deaths in European Union (0.00011) is larger than this value

of total deaths in Europe (0.00006)

 While the mode of total deaths in European Union is 0.00002, there is no mode

in Europe

To commence with, according to the outliers test below, it can be seen that the outliers which is the largest number of Europe so the best measures for this case is the median because if we utilize mean, its value will be affected a lot by the extreme value and

cannot cover all values On the other hand, there is no outlier in European Union;

therefore, mean is the most suitable measurement in this case as this can include as

many values as possible To sum up, it is obvious that the rate of total deaths in

European Union is approximately equal to the value of total deaths in Europe Hence, it also illustrates that European Union countries in particular and Europe countries in

general are being affected quite seriously by Covid-19 which lead to uncontrolled the number of deaths Acknowledging this, according to World Health Organization, this rate of total deaths will influence not only the economy but also significantly impact the environment; hence, the Europe group and the European Union are putting in place

restoration policies to bring natural life back (World Health Organization 2020)

 The range of total deaths in European Union is less than the range of total deaths

in Europe (0.00071 and 0.00313 respectively)

 The Interquartile of total deaths in European Union is greater than this value of total deaths in Europe (0.00030 and 0.00014 respectively)

 The variance of total deaths in European Union is lower than the variance of totaldeaths in Europe (0.00000004 and 0.0000005 respectively)

 The standard deviation of total deaths in European Union is bigger than this value of total deaths in Europe (0.00020 and 0.00069 respectively)

 The coefficient variation of total deaths in European Union is likely less than the coefficient variation of total deaths in Europe (101% and 258% respectively)

There is no doubt that Interquartile is the best measurement in this case To begin with, if

we take advantage of range, it just includes two values which cannot cover every value of data in European Union and Europe Secondly, because of the outlier in Europe, variance and standard deviation formulas which include the value of mean are not suitable in this case To conclude, it claims that Interquartile is better than other measures because utilizing Interquartile, we don’t need to worry about extreme value and avoid outlier

Box and Whisker Plot:

Figure 4: the box and whiskers plot of total deaths in European Union (Unit: per million population)

Figure 5: the box and whiskers plot of total deaths in Europe (Unit: per million population)

Left side >,<,= Right side Result

Median to extreme 0.00009 < 0.00062 Right-skewedFigure 6: Box and Whisker Plot of total deaths in European Union (Unit: per million population).

Left side >,<,= Right side Result

PART 3: Multiple Regression

Region A: European Union

H0: b2 = 0 (total death and average temperature have no relationship)

H1: b2 # 0 (total death and average temperature have a relationship)

From the result of excel: t-test value is equal to -0.312 and value is 0.759 Therefore, value is greater than 0.05, so we accept the null hypothesis In conclusion, total death and average temperature factor have no relationship

p-Because the p-value of Average temperature (X2) is largest and greater than 0.05, hence, weneed to eliminate Average temperature (X2)

o Medical doctor (X5)

H0: b5 = 0 (total death and medical doctor have no relationship)

H1: b5 # 0 (total death and medical doctor have a relationship)

From the excel: t-test value is -1.495 as well as p-value is equal to 0.156, as a result, p-value

is more than 0.05 Therefore, we accept the null hypothesis, which means total death and average temperature factor have no relationship Additionally, because the p-value of a medical doctor (X5) is largest and greater than 0.05, we eliminate the medical doctor factor (X5)

o Average Rainfall (X1)

H0: b1 = 0 (total death and average rainfall have no relationship)

H1: b1 # 0 (total death and average rainfall have relationship)

The result of excel shows that t-test value is equal to 1.449 and p-value is equal to 0.1665 Therefore, p-value is greater than 0.05, so we accept the null hypothesis Consequently, totaldeath and average temperature have no relationship In addition, because the p-value of Average rainfall (X1) is largest and more than 0.05, we eliminate the average rainfall factor (X1)

o Population (X4)

H0: b4 = 0 (total death and population have no relationship)

H1: b4 # 0 (total death and population have a relationship)

From the excel, t-test value is 1.595 as well as p-value is 0.129 From that, p-value is likely more than 0.05, we have to accept the null hypothesis Therefore, total death and population factor still have no relationship As a result, the p-value of population (X4) is larger than 0.05, we also eliminate the population factor (X4)

o Hospital beds (X3)

H0: b3 = 0 (total death and hospital beds have no relationship)

H1: b3 # 0 (total death and hospital beds have a relationship)

The result of excel shows that t-test value is equal to -2.046 and p-value is 0.056 In

comparison, p-value is greater than 0.05, so we accept the null hypothesis again For that reason, total death and hospital beds have no relationship

In conclusion:

As the final regression output of European Union, because the p-value of the last one, hospital beds (X3) is still greater than 0.05, hospital beds factor is insignificant independent variable As a result, we do not have any factors influencing the total number of deaths because of Covid-19 in 20 mentioned countries of European Union

 Regression function:

Because of no factor impacting the total number of deaths due to Covid-19, there is a function: expected total number of deaths owing to Covid-19 is equal to estimate of the regression intercept

Y(hat) = b0Estimated total deaths = 0.0004

c) Regression coefficient of the significant independent variables

b0 = 0.0004: The estimated total number of deaths because of Covid-19 in 20 countries of

European Union is equal to 0.0004 per million population This means that 20 nations above

will increase the number of deaths owing to Coronavirus, about 400 people per million people in total 20 countries, without any factors mentioned in the report This case is

nonsense because there are definitely reasons why Covid-19 deaths still steadily rise in thesecountries Therefore, the case can be make-sense if there are other factors impacting the change of this serious issue

d) The coefficient of determination

R Square = 13%: The result of the regression test is more powerful when the value of

R-Square is higher From that, the coefficient of determination is equal to 13%, which means that 13% of the change in expected total number of Covid-19 deaths can be explained by thechanges in these factors However, there are no factors in this case as well as the R-Square

is just 13% that is too small In addition, as the figure, the line chart illustrated that there was a dramatic reduction in the amount of deaths from Covid-19 in European Union

Therefore, we still have 87% for either mentioned causes or other ones not included in the report affecting the positive shift in the number of Covid-19 deaths The member States such as France, Germany, and Czechia supported medical equipment for Italy when Italy achieved the highest amount of coronavirus infected people (Anonymous 2020) From that,

it is clearly seen that the “solidarity clause” is an important factor influencing the decrease

of Covid-19 cases (Purnhagen, K.P, De Ruijter, Anniek, Flear, M.L, Hervey, T.K and Herwig, A, 2020)

H0: b4 = 0 (total death and population have no relationship)

H1: b4 # 0 (total death and population have a relationship)

From the excel, t-test value is equal to 0.631 and p-value is 0.538 In specific, p-value is likely more than 0.05, we have to accept the null hypothesis Therefore, total death and population factor have no relationship Additionally, because the p-value of population (X4)

is largest and greater than 0.05, we eliminate population factor (X4)

o Average Rainfall (X1)

H0: b1 = 0 (total death and average rainfall have no relationship)

H1: b1 # 0 (total death and average rainfall have a relationship)

From the regression analysis in excel, t-test value is -3.305 as well as p-value is equal to 0.008, hence, p-value is less than 0.05 Because of that, we reject the null hypothesis, and conclude total death and average rainfall (X1) have a relationship

o Average Temperature (X2)

H0: b2 = 0 (total death and average temperature have no relationship)

H1: b2 # 0 (total death and average temperature have a relationship)

From the calculation of excel, t-test value is equal to 2.142 and p-value is 0.049, so p-value

is likely less than 0.05 As a result, we reject the null hypothesis, and there is a relationship between total death and average temperature (X2)

o Hospital beds (X3)

H0: b3 = 0 (total death and hospital beds have no relationship)

H1: b3 # 0 (total death and hospital beds have a relationship)

From the excel, t-test value is -4.378 and p-value is equal to 0.0005, for that, p-value is smaller than 0.05 We reject the null hypothesis, and conclude that total death and hospital beds (X3) have a relationship

o Medical doctor (X5)

H0: b5 = 0 (total death and medical doctor have no relationship)

H1: b5 # 0 (total death and medical doctor have a relationship)

From the excel, t-test value is -1.495 as well as p-value is 0.156 For that, p-value is also smaller than 0.05, so we reject the null hypothesis Therefore, total death and average temperature (X5) have a relationship

 Multiple regression function:

MRF: Y(hat) = b0 + b1*X1 + b2*X2 + b3*X3 + b4*X5

Y(hat) = -0.00041 - 0.0000041*X1 + 0.0000596*X2 - 0.0000182*X3 + 0.0000369*X5

MRF Word: Estimated total deaths = -0.00041 - 0.0000041*average rainfall +

0.0000596*average temperature - 0.0000182*hospital beds + 0.0000369*medical doctorsY(hat): estimated total number of deaths (per million population) due to Covid-19 from April 01 to July 31, 2020

X1: average rainfall (in mm)

X2: average temperature (in celsius)

X3: hospital beds (per 10,000 people)

X5: medical doctors (per 10,000)

c) Regression coefficient of the significant independent variables

b0 = - 0.00041 illustrates that the total number of deaths of 20 countries in Europe will

increase to - 0.00041 per million population which is about 410 people per million

population without any factors mentioned in the report However this case is nonsense

because there are still some reasons that impacts on the total number of deaths as the result

of Covid-19

b1 = - 0.0000041 shows that the total number of deaths will increase approximately 4

people per million population when the average rainfall in mm increases 1 unit

b2 = 0.0000596 indicates that the total number of deaths will decrease over nearly 59 people

per million population when the average temperature in celsius increases 1 unit

b3 = - 0.0000182 illustrates the total number of deaths will increase around 18 people per

million population when hospital beds per 10,000 people raises 1 unit

b5 = 0.0000369 shows that the total number of deaths will reduce to about 37 persons per

million population when medical doctors per 10,000 climb 1 unit

d) The coefficient of determination

R square = 76%

It is obvious that the coefficient of determination is equal to 76% which means that 76% of the change in total number of Covid-19 deaths can be explained by the changes in these factors which are the average rainfall, the average temperature, hospital beds and medical doctors Hence, we still have 24% for other factors which not included in report According

to recent study, if the value of R square is closer to 1, the more accurate the result of the test will be higher Specifically, It can be seen that the value of R square is equal to

0,768130625 which nearly 1; hence, we can conclude that R square in this case is powerful for our test

PART 4: Conclusion

First of all, after running the regression test for 2 different regions: the European Union and Europe, there is an obvious difference between both regions Europe includes 4 factors, for instance, the average rainfall (in mm), the average temperature (in celsius), hospital beds (per 10,000 people), and medical doctors (per 10,000) influencing the total number of deathsfrom Covid-19 On the other hand, there are not any factors which are included in the report

to affect the change of total coronavirus deaths in the European Union

European Union

In analysed 20 nations of the region European Union, as the line chart illustrates that the event of this case is the reduction in the total number of deaths due to the spread of

coronavirus, which is good news for human health However, the final model of the

regression test shows that these mentioned factors do not influence the estimated total deaths From my research, the “solidary clause” is one of reasons to explain why the amount

of Covid-19 deaths goes down from 1st April to 31st July The member States cooperate andsupport each other to improve the health aspect via the Health Security Committee in Union law (Purnhagen, K.P, De Ruijter, Anniek, Flear, M.L, Hervey, T.K and Herwig, A, 2020) For example, Italy reached the highest number of Covid-19 deaths, compared to other nations collected in the report Additionally, based on Italy baseline scenario forecast, (Euromonitor Macro Model, 2020), Italy’s real GDP significantly decreased to minus number, -11.95% growth in 2020 because of the outbreak of coronavirus In 2021, it is predicted that real GDP will increase by 6.28% growth with the 41-51% of estimated

probability (figure 27) Thanks to the European solidarity, Germany delivered 300

ventilators as well as Czechia provided 10000 protective suits for Italy to support patients

and protect doctors, France also sent a million face masks (Anonymous 2020) Luckily, Italian scientists successfully discovered the Covid-19 vaccines to prevent the spread of this dangerous virus (Translated By Contentengine, L.L.C., 2020)

Europe

Based on the final model of the regression test we mentioned above, there are 4 factors which are the average rainfall, the average temperature, hospital beds and medical doctors impact significantly on the total number of deaths To commence with, according to recent studies, extreme weather conditions which are the average climatic temperatures and the average rainfall play an important role in the rapid spread of viruses Specifically, if the average rainfall is lower, it means that the death rate from covid-19 increases, however, if the average temperature increases, the number of deaths decreases because virus corona cannot live in hot climates It is a biological catalyst for the interaction between covid-19 and the human respiratory system (Mesay Moges Menebo, 2020) In addition, due to our collected data, while the average rainfall for the 20 countries in Europe is estimated at 97,72854416 mm, the average temperature for those countries is 9,647996515 celsius Thus,

as the result of regression coefficient, average rainfall can increase death rates while averagetemperatures can reduce death rates On the other hand, a smaller supply of hospital beds leads to higher mortality rates that induce governments to take more stringent closure measures that have greater economic costs (Nathan Sussman, 2020) Hence, in this case, the total number of hospital beds is equal to 956,16 per 10,000 population which leads to the higher mortality rates because of covid-19 In contrast, doctors' human resources also play

an important role in the fight with Covid-19 because if the number of doctors increases, patients who suffer from coronavirus infection have an increased chance of being treated (Civantos, 2020) Therefore, 20 countries in Europe have a total of 742,11 (per 10,000) medical doctors which means that the mortality rate decreases

To conclude, owing to the descriptive measurements part, it is suggested that the rate of total deaths in European Union is approximately equal to the value of total deaths in Europe.However, based on what we mentioned in part 3 and part 4, there is no doubt that European Union countries are higher impacted than Europe countries due to the Covid-19 pandemic Itcan be said that since the outbreak in Italy and began to spread to all European Union countries, this bloc has reacted relatively passively Acknowledging this, the European Union (EU) has decided to close all land, sea and air borders of this bloc within 30 days; before that was the spending 25 billion euros to respond to the crisis Therefore, European Commission and the World Health Organisation to strengthen coronavirus treatments in global collaboration (European Commission 2020)

PART 5: Time Series

REGION A: EUROPEAN UNION

1 Output and formula of linear (LIN), quadratic (QUA) and exponential (EXP) trend model.

1.1 Linear Trend Model (LIN)

a Provide the regression output in the report

Figure 10: Output of time series for Linear Trend Model in European Union.

b Provide the formula of trend model

1.2 Quadratic Trend Model (QUA)

a Provide the regression output in the report

Figure 11: Output of time series for Quadratic in European Union

b Provide the formula of the trend model

1.3 Exponential Trend Model (EXP)

a Provide the regression output in the report