FinalProject_Code_v1

######################################################################################################################
                                                #Import packages
######################################################################################################################

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns

######################################################################################################################
                                                #Import Data from Excel
######################################################################################################################

#1. Import data

file='WDIW Dataset.xlsx'

worldData_df=pd.read_excel(file, index_col=0)

#CountryCode             = 'CountryCode'
CountryName             = 'CountryName'
HultRegion              = 'Hult Region'
CoolName                = 'Cool Name'
ElectricityAccessTotal  = 'ElectricityAccessTotal'
ElectricityRuralAccess  = 'ElectricityRuralAccess'
ElectricityUrbanAccess  = 'ElectricityUrbanAccess'
AdolescentFertility     = 'AdolescentFertility'
AgeDependencyTotal      = 'AgeDependencyTotal'
AgeDependencyOld        = 'AgeDependencyOld'
AgeDependencyYoung      = 'AgeDependencyYoung'
GDP_AgriForFish         = 'GDP_PrimaryActivities'
ArmedForces             = 'ArmedForces'
BirthRate               = 'BirthRate'
BirthSkilledStaff       = 'BirthSkilledStaff'
DeathRate               = 'DeathRate'
EducationPHD            = 'EducationPHD'
EducationBachelor       = 'EducationBachelor'
EducationMaster         = 'EducationMaster'
EducationLowerSecondary = 'EducationLowerSecondary'
EducationPostSecondary  = 'EducationPostSecondary'
EducationPrimary        = 'EducationPrimary'
EducationTertiary       = 'EducationTertiary'
EducationUpperSecondary = 'EducationUpperSecondary'
EmploymentAgriculture   = 'EmploymentAgriculture'
EmploymentIndustry      = 'EmploymentIndustry'
EmploymentServices      = 'EmploymentServices'
FertilityRate           = 'FertilityRate'
GDP                     = 'GDP'
GDPGrowth               = 'GDPGrowth'
GINI                    = 'GINI'
ExpendEducation         = 'ExpendEducation'
IncomeFourth            = 'IncomeFourth'
IncomeFifth             = 'IncomeFifth'
IncomeFirst             = 'IncomeFirst'
IncomeSecond            = 'IncomeSecond'
IncomeThird             = 'IncomeThird'
GDP_Industry            = 'GDP_Industry'
LifeExpectancy          = 'LifeExpectancy'
LiteracyAdult           = 'LiteracyAdult'
LiteracyYouth           = 'LiteracyYouth'
GDP_Merchandise         = 'GDP_Merchandise'
ExpendMilitary          = 'ExpendMilitary'
MobileSubs              = 'MobileSubs'
PovertyByHealthcare     = 'PovertyByHealthcare'
Population0_14          = 'Population0_14'
Population15_64         = 'Population15_64'
Population65            = 'Population65+'
PopulationDensity       = 'PopulationDensity'
PopulationGrowth        = 'PopulationGrowth'
PopulationLargestCity   = 'PopulationLargestCity'
PopulationSlums         = 'PopulationSlums'
PopulationFemale        = 'PopulationFemale'
PopulationMale          = 'PopulationMale'
PopulationTotal         = 'PopulationTotal'
Poverty_1               = 'Poverty_1.90'
Poverty_3               = 'Poverty_3.20'
Poverty_NationalLine    = 'Poverty_NationalLine'
HIV                     = 'HIV'
Undernourishment        = 'Undernourishment'
Underweight             = 'Underweight'
PopulationRural         = 'PopulationRural'
GDP_Services            = 'GDP_Services'
SurfaceArea             = 'SurfaceArea'
TaxRevenue              = 'TaxRevenue'
PopulationUrban         = 'PopulationUrban'
PopulationUrbanGrowth   = 'PopulationUrbanGrowth'

newColumns=[#CountryCode,
            CountryName, HultRegion, CoolName, ElectricityAccessTotal, ElectricityRuralAccess,\
            ElectricityUrbanAccess, AdolescentFertility, AgeDependencyTotal, AgeDependencyOld, AgeDependencyYoung, GDP_AgriForFish,\
            ArmedForces, BirthRate, BirthSkilledStaff, DeathRate, EducationPHD, EducationBachelor, \
            EducationMaster, EducationLowerSecondary, EducationPostSecondary, EducationPrimary, EducationTertiary, EducationUpperSecondary,\
            EmploymentAgriculture, EmploymentIndustry, EmploymentServices, FertilityRate, GDP, GDPGrowth,\
            GINI, ExpendEducation, IncomeFourth, IncomeFifth, IncomeFirst, IncomeSecond,\
            IncomeThird, GDP_Industry, LifeExpectancy, LiteracyAdult, LiteracyYouth, GDP_Merchandise,\
            ExpendMilitary, MobileSubs, PovertyByHealthcare, Population0_14, Population15_64, Population65,\
            PopulationDensity, PopulationGrowth, PopulationLargestCity, PopulationSlums, PopulationFemale, PopulationMale,\
            PopulationTotal, Poverty_1, Poverty_3, Poverty_NationalLine, HIV, Undernourishment,\
            Underweight, PopulationRural, GDP_Services, SurfaceArea, TaxRevenue, PopulationUrban,\
            PopulationUrbanGrowth]

worldData_df.columns = newColumns

#2. Creation of the subset for Eastern Europe

teamData_df=worldData_df[worldData_df['Cool Name']=='Skids & Mudflap']

#3. Eliminating Andorra and San Marino

teamDataCleaned_df=teamData_df.copy()

teamDataCleaned_df=teamDataCleaned_df.drop(labels=['AND','SMR'])

######################################################################################################################
                                                #Data Analysis
######################################################################################################################

#1. Adolescent Fertility

print(teamData_df['AdolescentFertility'].describe())
print('\n')
print(worldData_df['AdolescentFertility'][worldData_df['CountryName']=='World'])

#Histogram with missing values
teamData_df.hist(column=['AdolescentFertility'], bins='sturges')
plt.title('Eastern Europe')
plt.xlabel('Adolescent Fertility')
plt.show()

#The distribution is somewhat normal so we will use the mean for completing Kosovo

adolescentFertility_mean = teamDataCleaned_df['AdolescentFertility'].mean()
teamDataCleaned_df[['AdolescentFertility']] = teamDataCleaned_df[['AdolescentFertility']].fillna(adolescentFertility_mean)

print(teamDataCleaned_df['AdolescentFertility'].describe())

print(worldData_df['AdolescentFertility'][worldData_df['CountryName']=='World'])

sns.distplot(teamDataCleaned_df['AdolescentFertility'], bins='sturges')
plt.title('Eastern Europe')
plt.xlabel("AdolescentFertility")
plt.show()

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['AdolescentFertility'])
plt.title('AdolescentFertility')
plt.xticks(rotation=90)
plt.show()

#2. Age Dependecy Total

#We use the median of balcanic countries to fill Kosovo

ageDependency_median=teamDataCleaned_df.loc[['BIH','HRV','SVN','ALB','SRB','MKD','MNE'],['AgeDependencyTotal']].median()

teamDataCleaned_df[['AgeDependencyTotal']] = teamDataCleaned_df[['AgeDependencyTotal']].fillna(ageDependency_median)

#Description of the variable and World average

print(teamDataCleaned_df['AgeDependencyTotal'].describe())
print('\n')
print(worldData_df['AgeDependencyTotal'][worldData_df['CountryName']=='World'])

#Bar chat

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['AgeDependencyTotal'])
plt.title('Age Dependency Total')
plt.xticks(rotation=90)
plt.show()

#Histogram for Eastern Europe

teamDataCleaned_df.hist(column=['AgeDependencyTotal'], bins='sturges')
plt.title('Eastern Europe')
plt.xlabel('Age Dependency Total')
plt.show()

#Histogram for Western Europe

worldData_df['AgeDependencyTotal'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('Age Dependency Total')
plt.show()

#Histogram for the Rest of the World (Complement for Eastern Europe only)

worldData_df['AgeDependencyTotal'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('Age Dependency Total')
plt.show()

#3. Age Dependency Old

#We use the median of balcanic countries to fill Kosovo

ageDependencyOld_median=teamDataCleaned_df.loc[['BIH','HRV','SVN','ALB','SRB','MKD','MNE'],['AgeDependencyOld']].median()

teamDataCleaned_df[['AgeDependencyOld']] = teamDataCleaned_df[['AgeDependencyOld']].fillna(ageDependencyOld_median)

#Description of the variable and World average

print(teamDataCleaned_df['AgeDependencyOld'].describe())
print('\n')
print(worldData_df['AgeDependencyOld'][worldData_df['CountryName']=='World'])

#Bar chat

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['AgeDependencyOld'])
plt.title('Age Dependency Old')
plt.xticks(rotation=90)
plt.show()

#Histogram for Eastern Europe

teamData_df.hist(column=['AgeDependencyOld'], bins='sturges')
plt.title('Eastern Europe')
plt.xlabel('Age Dependency Old')
plt.show()

#Histogram for Western Europe

worldData_df['AgeDependencyOld'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('Age Dependency Old')
plt.show()

#Histogram for the Rest of the World (Complement for Eastern Europe only)

worldData_df['AgeDependencyOld'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('Age Dependency Old')
plt.show()

#4. Age Dependency Young

#We use the median of balcanic countries to fill Kosovo

ageDependencyOld_median=teamDataCleaned_df.loc[['BIH','HRV','SVN','ALB','SRB','MKD','MNE'],['AgeDependencyYoung']].median()

teamDataCleaned_df[['AgeDependencyYoung']] = teamDataCleaned_df[['AgeDependencyYoung']].fillna(ageDependencyOld_median)

#Description of the variable and World average

print(teamDataCleaned_df['AgeDependencyYoung'].describe())
print('\n')
print(worldData_df['AgeDependencyYoung'][worldData_df['CountryName']=='World'])

#Bar chat

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['AgeDependencyYoung'])
plt.title('Age Dependency Young')
plt.xticks(rotation=90)
plt.show()

#Histogram for Eastern Europe

teamData_df.hist(column=['AgeDependencyYoung'], bins='sturges')
plt.title('Eastern Europe')
plt.xlabel('Age Dependency Young')
plt.show()

#Histogram for Western Europe

worldData_df['AgeDependencyYoung'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('Age Dependency Young')
plt.show()

#Histogram for the Rest of the World (Complement for Eastern Europe only)

worldData_df['AgeDependencyOld'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('Age Dependency Old')
plt.show()

#5. GDP Primary Activities

#We fill the missing value of Poland with data from CIA

teamDataCleaned_df.loc['POL','GDP_PrimaryActivities']=2.7

#Description of the variable and World average
print(teamDataCleaned_df['GDP_PrimaryActivities'].describe())
print('\n')
print(worldData_df['GDP_PrimaryActivities'][worldData_df['CountryName']=='World'])

#Bar chart
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDP_PrimaryActivities'])
plt.title('GDP_PrimaryActivities')
plt.xticks(rotation=90)
plt.show()

#Histogram
teamDataCleaned_df.hist(column=['GDP_PrimaryActivities'])
plt.show()

#Clear outlier: Albania

#6. Birth Rate

#Description of the variable and World average

print(teamDataCleaned_df['BirthRate'].describe())
print('\n')
print(worldData_df['BirthRate'][worldData_df['CountryName']=='World'])

#Bar Chart with raw values

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['BirthRate'])
plt.title('BirthRate')
plt.xticks(rotation=90)
plt.show()

#We create a new variable that converts the birth rate to proportion of the population

teamDataCleaned_df['BirthRateProportion']=teamDataCleaned_df['BirthRate']*1000*100/teamDataCleaned_df['PopulationTotal']

#Bar chart with the new variable
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['BirthRateProportion'])
plt.xticks(rotation=90)
plt.title('BirthRateProportion')
plt.show()

#7. Death Rate

#Description of the variable and World average

print(teamDataCleaned_df['DeathRate'].describe())
print('\n')
print(worldData_df['DeathRate'][worldData_df['CountryName']=='World'])

#We create a new variable that converts the death rate to proportion of the population

teamDataCleaned_df['DeathRateProportion']=teamDataCleaned_df['DeathRate']*1000*100/teamDataCleaned_df['PopulationTotal']

#Bar Chart with raw values

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['DeathRate'])
plt.title('DeathRate')
plt.xticks(rotation=90)
plt.show()

#Bar chart with the new variable

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['DeathRateProportion'])
plt.title('DeathRateProportion')
plt.xticks(rotation=90)
plt.show()

#8. Population Total

#Description of the variable and World average

print(teamData_df['PopulationTotal'].describe())
print('\n')
print(worldData_df['PopulationTotal'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamData_df.index, height=teamData_df['PopulationTotal'])
plt.xticks(rotation=90)
plt.show()

#Histogram
teamData_df.hist(column=['PopulationTotal'], bins='sturges')
plt.show()

#9. Male Proportion

#We convert male population in another variable called Male Proportion

teamDataCleaned_df['MaleProportion']=pd.DataFrame(teamData_df['PopulationMale']/teamData_df['PopulationTotal'])

#We complete the missing value for Kosovo with the median

maleProportion_median = teamDataCleaned_df['MaleProportion'].median()
teamDataCleaned_df[['MaleProportion']] = teamDataCleaned_df[['MaleProportion']].fillna(maleProportion_median)

#We complete Male Population for Kosovo

teamDataCleaned_df.loc['XKX','PopulationMale']=round(teamDataCleaned_df.loc['XKX','PopulationTotal']*teamDataCleaned_df.loc['XKX','MaleProportion'])
      
#Bar chart for Male Proportion

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['MaleProportion'])
plt.xticks(rotation=90)
plt.show()

#Histogram for Male Proportion
teamDataCleaned_df.hist(column=['MaleProportion'], bins='sturges')
plt.show()

#10. Female Proportion

#We convert female population in another variable called Female Proportion

teamDataCleaned_df['FemaleProportion']=pd.DataFrame(teamData_df['PopulationFemale']/teamData_df['PopulationTotal'])

#We complete the missing value for Kosovo with the median

femaleProportion_median = teamDataCleaned_df['FemaleProportion'].median()
teamDataCleaned_df[['FemaleProportion']] = teamDataCleaned_df[['FemaleProportion']].fillna(femaleProportion_median)

#We complete Female Population for Kosovo

teamDataCleaned_df.loc['XKX','PopulationFemale']=round(teamDataCleaned_df.loc['XKX','PopulationTotal']*teamDataCleaned_df.loc['XKX','FemaleProportion'])

#Bar chart for Female Proportion

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['FemaleProportion'])
plt.xticks(rotation=90)
plt.show()

#Histogram for Female Proportion
teamDataCleaned_df.hist(column=['FemaleProportion'], bins='sturges')
plt.show()

#11. Population Slums

#Description of the variable and World average

print(teamDataCleaned_df['PopulationSlums'].describe())
print('\n')
print(worldData_df['PopulationSlums'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationSlums'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['PopulationSlums'], bins='sturges')
plt.show()

#12. Population Largest City

#We fill the missing values with data found on research and the median for Kosovo

teamDataCleaned_df.loc['MLT','PopulationLargestCity']= 5
teamDataCleaned_df.loc['MNE','PopulationLargestCity']= 42.8
teamDataCleaned_df.loc['SVN','PopulationLargestCity']= 25.7

Kosovomedcity = teamDataCleaned_df['PopulationLargestCity'].median()
teamDataCleaned_df[['PopulationLargestCity']] = teamDataCleaned_df[['PopulationLargestCity']].fillna(Kosovomedcity)

#Description of the variable and World average

print(teamDataCleaned_df['PopulationLargestCity'].describe())
print('\n')
print(worldData_df['PopulationLargestCity'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationLargestCity'])
plt.xticks(rotation=90)
plt.show()

#Histogram
teamDataCleaned_df.hist(column=['PopulationLargestCity'], bins='sturges')
plt.show()

#13. Employment Agriculture

#We fill the missing value with the median

EmploymentAgriculture_median = teamDataCleaned_df['EmploymentAgriculture'].median()
teamDataCleaned_df[['EmploymentAgriculture']] = teamDataCleaned_df[['EmploymentAgriculture']].fillna(EmploymentAgriculture_median)

#Description of the variable and World average

print(teamDataCleaned_df['EmploymentAgriculture'].describe())
print('\n')
print(worldData_df['EmploymentAgriculture'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['EmploymentAgriculture'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['EmploymentAgriculture'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World

worldData_df['EmploymentAgriculture'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('EmploymentAgriculture')
plt.show()

worldData_df['EmploymentAgriculture'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('EmploymentAgriculture')
plt.show()

#14. Employment Industry

#We fill the missing value with the median

EmploymentIndustry_median = teamDataCleaned_df['EmploymentIndustry'].median()
teamDataCleaned_df[['EmploymentIndustry']] = teamDataCleaned_df[['EmploymentIndustry']].fillna(EmploymentIndustry_median)

#Description of the variable and World average

print(teamDataCleaned_df['EmploymentIndustry'].describe())
print('\n')
print(worldData_df['EmploymentIndustry'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['EmploymentIndustry'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['EmploymentIndustry'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World

worldData_df['EmploymentIndustry'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('EmploymentIndustry')
plt.show()

worldData_df['EmploymentIndustry'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('EmploymentIndustry')
plt.show()

#15. Employment Services

#We fill the missing value with the median

EmploymentServices_median = teamDataCleaned_df['EmploymentServices'].median()
teamDataCleaned_df[['EmploymentServices']] = teamDataCleaned_df[['EmploymentServices']].fillna(EmploymentServices_median)

#Description of the variable and World average

print(teamDataCleaned_df['EmploymentServices'].describe())
print('\n')
print(worldData_df['EmploymentServices'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['EmploymentServices'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['EmploymentServices'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World
worldData_df['EmploymentServices'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('EmploymentServices')
plt.show()

worldData_df['EmploymentServices'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('EmploymentServices')
plt.show()

#16. Employment Distribution (graph only)

#Define graph size

plt.figure(figsize =(12, 8))

#Create Agriculture bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['EmploymentAgriculture'], color='#b5ffb9')
#Create Industry Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['EmploymentIndustry'], color='#f9bc86', edgecolor='white', left=teamDataCleaned_df['EmploymentAgriculture'])
#Create Services Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['EmploymentServices'], color='#a3acff', edgecolor='white', left=(teamDataCleaned_df['EmploymentAgriculture']+teamDataCleaned_df['EmploymentIndustry']))

#Set legend

legend=['Agriculture','Industry','Services']
plt.legend(legend)

plt.show()

#17. Fertility Rate

#Description of the variable and World average

print(teamDataCleaned_df['FertilityRate'].describe())
print('\n')
print(worldData_df['FertilityRate'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['FertilityRate'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['FertilityRate'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World

worldData_df['FertilityRate'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('FertilityRate')
plt.show()

worldData_df['FertilityRate'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('FertilityRate')
plt.show()

#18. GDP and GDP per Capita

#Calculation of GDP per Capita

teamDataCleaned_df['GDPCapita']=pd.DataFrame(teamData_df['GDP']/teamData_df['PopulationTotal'])

#Bar chart for GDP

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDP'])
plt.title('GDP')
plt.xticks(rotation=90)
plt.show()

#Histogram for GDP

teamDataCleaned_df.hist(column=['GDP'], bins='sturges')
plt.title('GDP')
plt.show()

#Bar chart for GDP per Capita

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDPCapita'])
plt.title('GDP per Capita')
plt.xticks(rotation=90)
plt.show()

#Histogram for GDP

teamDataCleaned_df.hist(column=['GDPCapita'], bins='sturges')
plt.title('GDP per Capita')
plt.show()

#19. GDP Growth

#Description of the variable and World average

print(teamDataCleaned_df['GDPGrowth'].describe())
print('\n')
print(worldData_df['GDPGrowth'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamData_df.index, height=teamData_df['GDPGrowth'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamData_df.hist(column=['GDPGrowth'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World

worldData_df['GDPGrowth'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('GDPGrowth')
plt.show()

worldData_df['GDPGrowth'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('GDPGrowth')
plt.show()

#20. GDP Services

#We complete the data for Poland found on research

teamDataCleaned_df.loc['POL','GDP_Services']= 58.9

#Description of the variable and World average

print(teamDataCleaned_df['GDP_Services'].describe())
print('\n')
print(worldData_df['GDP_Services'][worldData_df['CountryName']=='World'])


#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDP_Services'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['GDP_Services'], bins='sturges')
plt.show()

#Histogram for Western Europe

worldData_df['GDP_Services'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('GDP_Services')
plt.show()

#21. Armed Forces

#Kosovo officially announced an army on 2018, therefore it should be zero

teamDataCleaned_df['ArmedForces'] = teamDataCleaned_df['ArmedForces'].fillna(0)

#Description of the variable and World average

print(teamDataCleaned_df['ArmedForces'].describe())
print('\n')
print(worldData_df['ArmedForces'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.title('Armed Forces - % of Total labor force')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['ArmedForces'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot
teamDataCleaned_df.hist(column=['ArmedForces'], bins='sturges')
plt.show()

#Seaborn plot
sns.distplot(teamDataCleaned_df['ArmedForces'], bins='sturges')
plt.title("Armed Forces - % of total labor force - Seaborn")
plt.show()

#22. GDP Industry

#First we complete Poland's GDP Industry from research

teamDataCleaned_df.loc['POL','GDP_Industry']= 38.4

#Description of the variable and World average

print(teamDataCleaned_df['GDP_Industry'].describe())
print('\n')
print(worldData_df['GDP_Industry'][worldData_df['CountryName']=='World'])

#X-bar plot - Filled the gaps

plt.title('Industry (% of GDP)')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDP_Industry'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['GDP_Industry'], bins='sturges')
plt.show()

#Seaborn plot

sns.distplot(teamDataCleaned_df['GDP_Industry'], bins='sturges')
plt.title("Industry (% of GDP) - Seaborn")
plt.show()

#23. GDP Composition (graph only)

#Define graph size

plt.figure(figsize =(12, 8))

#Create Agriculture bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['GDP_PrimaryActivities'], color='#b5ffb9')
#Create Industry Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['GDP_Industry'], color='#f9bc86', edgecolor='white', left=teamDataCleaned_df['GDP_PrimaryActivities'])
#Create Services Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['GDP_Services'], color='#a3acff', edgecolor='white', left=(teamDataCleaned_df['GDP_PrimaryActivities']+teamDataCleaned_df['GDP_Industry']))

#Set legend

legend=['Agriculture','Industry','Services']
plt.legend(legend)
plt.title('GDP Composition')

plt.show()

#24. Life Expectancy

#Description of the variable and World average

print(teamDataCleaned_df['LifeExpectancy'].describe())
print('\n')
print(worldData_df['LifeExpectancy'][worldData_df['CountryName']=='World'])

#X-bar plot

plt.title('Life Expectancy at Birth (in years)')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['LifeExpectancy'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot
teamDataCleaned_df.hist(column=['LifeExpectancy'], bins='sturges')
plt.show()

#Histogram for Western Europe

worldData_df['LifeExpectancy'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('LifeExpectancy')
plt.show()

#25. GDP Merchandise

#We fill the missing value for Kosovo with the mean of the lowest countries

GDPMerch_mean = teamDataCleaned_df.loc[['FIN','GRC','NOR','SWE'],'GDP_Merchandise'].mean()
teamDataCleaned_df['GDP_Merchandise'] = teamDataCleaned_df['GDP_Merchandise'].fillna(GDPMerch_mean)

#Description of the variable and World average

print(teamDataCleaned_df['GDP_Merchandise'].describe())
print('\n')
print(worldData_df['GDP_Merchandise'][worldData_df['CountryName']=='World'])

#X-bar plot

plt.title('Merchandise Trade (% of GDP)')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['GDP_Merchandise'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['GDP_Merchandise'], bins='sturges')
plt.show()

#Seaborn plot
sns.distplot(teamDataCleaned_df['GDP_Merchandise'], bins='sturges')
plt.title("Merchandise Trade (% of GDP) - Seaborn")
plt.show()

#26. Military Expenditures

#Description of the variable and World average

print(teamDataCleaned_df['ExpendMilitary'].describe())
print('\n')
print(worldData_df['ExpendMilitary'][worldData_df['CountryName']=='World'])

#X-bar plot

plt.title('Military Expenditure( % of GDP)')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['ExpendMilitary'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['ExpendMilitary'], bins='sturges')
plt.show()

#Seaborn plot
sns.distplot(teamDataCleaned_df['ExpendMilitary'], bins='sturges')
plt.title("Military Expenditure( % of GDP) - Seaborn")
plt.show()

#27. Mobile Subscribers

#We fill Kosovo with the median

mobileSubs_median = teamDataCleaned_df['MobileSubs'].median()
teamDataCleaned_df['MobileSubs'] = teamDataCleaned_df['MobileSubs'].fillna(mobileSubs_median)

#Description of the variable and World average

print(teamDataCleaned_df['MobileSubs'].describe())
print('\n')
print(worldData_df['MobileSubs'][worldData_df['CountryName']=='World'])

#X-bar plot

plt.title('Mobile Subcribers (per 100 people)')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['MobileSubs'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['MobileSubs'], bins='sturges')
plt.show()

#Seaborn plot

sns.distplot(teamDataCleaned_df['MobileSubs'], bins='sturges')
plt.title("Mobile Subcribers (per 100 people) - Seaborn")
plt.show()

#28. Poverty by Healthcare

#Description of the variable and World average

print(teamDataCleaned_df['PovertyByHealthcare'].describe())
print('\n')
print(worldData_df['PovertyByHealthcare'][worldData_df['CountryName']=='World'])

#X-bar plot
plt.title('Below Poverty Line < $3.10')
plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PovertyByHealthcare'])
plt.xticks(ticks=teamDataCleaned_df.index, labels=teamDataCleaned_df['CountryName'],rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['PovertyByHealthcare'], bins='sturges')
plt.show()

#Seaborn plot

sns.distplot(teamDataCleaned_df['PovertyByHealthcare'], bins='sturges')
plt.title("Below Poverty Line < $3.10 - Seaborn")
plt.show()

#29. Population Growth

#Description of the variable and World average

print(teamDataCleaned_df['PopulationGrowth'].describe())
print('\n')
print(worldData_df['PopulationGrowth'][worldData_df['CountryName']=='World'])


#X-bar plot

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationGrowth'])
plt.xticks(rotation=90)
plt.show()

#Histogram Plot

teamDataCleaned_df.hist(column=['PopulationGrowth'], bins='sturges')
plt.show()

#30. PopulationDensity

#Description of the variable and World average

print(teamDataCleaned_df['PopulationDensity'].describe())
print('\n')
print(worldData_df['PopulationDensity'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationDensity'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['PopulationDensity'], bins='sturges')
plt.show()

#31. Population65+

#We complete Kosovo with the median

population65_median = round(teamDataCleaned_df['Population65+'].median())
teamDataCleaned_df['Population65+'] = teamDataCleaned_df['Population65+'].fillna(population65_median)

#Description of the variable and World average

print(teamDataCleaned_df['Population65+'].describe())
print('\n')
print(worldData_df['Population65+'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['Population65+'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['Population65+'], bins='sturges')
plt.show()

#32. Population15_64

#We complete Kosovo with the median

population15_64_median = round(teamDataCleaned_df['Population15_64'].median())
teamDataCleaned_df['Population15_64'] = teamDataCleaned_df['Population15_64'].fillna(population15_64_median)

#Description of the variable and World average

print(teamDataCleaned_df['Population15_64'].describe())
print('\n')
print(worldData_df['Population15_64'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['Population15_64'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['Population15_64'], bins='sturges')
plt.show()

#33. Population0_14

#We complete Kosovo with the median

population0_14_median = round(teamDataCleaned_df['Population0_14'].median())
teamDataCleaned_df['Population0_14'] = teamDataCleaned_df['Population0_14'].fillna(population0_14_median)

#Description of the variable and World average

print(teamDataCleaned_df['Population0_14'].describe())
print('\n')
print(worldData_df['Population0_14'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['Population0_14'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['Population0_14'], bins='sturges')
plt.show()

#34. Population Age Distribution (graph only)

#Define graph size

plt.figure(figsize =(12, 8))

#Create 0-14 bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['Population0_14'], color='#b5ffb9')
#Create 15-64 Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['Population15_64'], color='#f9bc86', edgecolor='white', left=teamDataCleaned_df['Population0_14'])
#Create 65+ Bar
plt.barh(y=teamDataCleaned_df.index, width=teamDataCleaned_df['Population65+'], color='#a3acff', edgecolor='white', left=(teamDataCleaned_df['Population0_14']+teamDataCleaned_df['Population15_64']))

#Set legend

legend=['0 - 14','15 - 64','65+']
plt.legend(legend)
plt.title('Age Distribution')

plt.show()

#35. Undernourishment

#Kosovo is missing but most countries seem to report the same vaues for undernourishment

#Description of the variable and World average

print(teamDataCleaned_df['Undernourishment'].describe())
print('\n')
print(worldData_df['Undernourishment'][worldData_df['CountryName']=='World'])

#Bar Chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['Undernourishment'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['Undernourishment'], bins='sturges')
plt.show()

#36. Population Rural

#We fill Kosovo with the median

kosovo_median_PopulationRural= teamDataCleaned_df['PopulationRural'].median()
teamDataCleaned_df.loc['XKX','PopulationRural']=kosovo_median_PopulationRural

#Description of the variable and World average

print(teamDataCleaned_df['PopulationRural'].describe())
print('\n')
print(worldData_df['PopulationRural'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationRural'])
plt.xticks(rotation=90)

#Histogram

teamDataCleaned_df.hist(column=['PopulationRural'], bins='sturges')
plt.show()

#37. Surface Area

#Description of the variable and World average

print(teamDataCleaned_df['SurfaceArea'].describe())
print('\n')
print(worldData_df['SurfaceArea'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['SurfaceArea'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['SurfaceArea'], bins='sturges')
plt.show()

#38. Tax Revenue GDP

#Serbia, Montenegro and Kosovo have missing values and all were one country some years ago

#Description of the variable and World average

print(teamDataCleaned_df['TaxRevenue'].describe())
print('\n')
print(worldData_df['TaxRevenue'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['TaxRevenue'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['TaxRevenue'], bins='sturges')
plt.show()

#Adding histograms for Western Europe and the Rest of the World

worldData_df['TaxRevenue'][worldData_df['Cool Name']=='Jetfire'].hist(bins='sturges')
plt.title('Western Europe')
plt.xlabel('% Tax Revenue from GDP')
plt.show()

worldData_df['TaxRevenue'][worldData_df['CountryName']!='World'][worldData_df['Cool Name']!='Skids & Mudflap'].hist(bins='sturges')
plt.title('Rest of the World')
plt.xlabel('% Tax Revenue from GDP')
plt.show()

#39. Population Urban

#We fill Kosovo with the median

kosovo_median_PopulationUrban= teamDataCleaned_df['PopulationUrban'].median()
teamDataCleaned_df.loc['XKX','PopulationUrban']=kosovo_median_PopulationUrban

#Description of the variable and World average

print(teamDataCleaned_df['PopulationUrban'].describe())
print('\n')
print(worldData_df['PopulationUrban'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationUrban'])
plt.xticks(rotation=90)

#Histogram

teamDataCleaned_df.hist(column=['PopulationUrban'], bins='sturges')
plt.show()

#40. Population Urban Growth

#Description of the variable and World average

print(teamDataCleaned_df['PopulationUrbanGrowth'].describe())
print('\n')
print(worldData_df['PopulationUrbanGrowth'][worldData_df['CountryName']=='World'])

#Bar chart

plt.bar(x=teamDataCleaned_df.index, height=teamDataCleaned_df['PopulationUrbanGrowth'])
plt.xticks(rotation=90)
plt.show()

#Histogram

teamDataCleaned_df.hist(column=['PopulationUrbanGrowth'], bins='sturges')
plt.show()

######################################################################################################################
                                                #Correlation Matrix
######################################################################################################################

correlationData_df=teamDataCleaned_df.dropna(axis=1, thresh=15).corr().dropna(thresh=1).dropna(axis=1, thresh=15)
f = plt.figure(figsize=(19, 15))
plt.matshow(correlationData_df, fignum=f.number)
plt.xticks(range(correlationData_df.shape[1]), correlationData_df.columns, fontsize=14, rotation=90)
plt.yticks(range(correlationData_df.shape[1]), correlationData_df.columns, fontsize=14)
plt.grid(True)
cb = plt.colorbar()
cb.ax.tick_params(labelsize=14)