Covid Country Analysis¶

In [1]:
library(ggplot2)
library(hrbrthemes)
library(GGally)
library(corrgram)
library(visdat)
library(ISLR2)
library(e1071)
library(caTools)
library(hopkins)
#library(randomForest)

NOTE: Either Arial Narrow or Roboto Condensed fonts are required to use these themes.

      Please use hrbrthemes::import_roboto_condensed() to install Roboto Condensed and

      if Arial Narrow is not on your system, please see https://bit.ly/arialnarrow

Registered S3 method overwritten by 'GGally':
  method from   
  +.gg   ggplot2

Context

  • A new coronavirus designated 2019-nCoV was first identified in Wuhan, the capital of China's Hubei province
  • People developed pneumonia without a clear cause and for which existing vaccines or treatments were not effective.
  • The virus has shown evidence of human-to-human transmission
  • Transmission rate (rate of infection) appeared to escalate in mid-January 2020
  • As of 30 January 2020, approximately 8,243 cases have been confirmed
  • full_grouped.csv - Day to day country wise no. of cases (Has County/State/Province level data)
  • covid19clean_complete.csv - Day to day country wise no. of cases (Doesn't have County/State/Province level data)
  • countrywiselatest.csv - Latest country level no. of cases
  • day_wise.csv - Day wise no. of cases (Doesn't have country level data)
  • usacountywise.csv - Day to day county level no. of cases
  • worldometer_data.csv - Latest data from https://www.worldometers.info/
In [2]:
# Import datasets from csv

# Source: https://www.kaggle.com/datasets/imdevskp/corona-virus-report?resource=download&select=country_wise_latest.csv

# Online from github repo
df = read.csv('https://raw.githubusercontent.com/Giovo17/covid-country-analysis/main/country_wise_latest.csv')

# Local from disk
#df = read.csv("country_wise_latest.csv")
In [3]:
head(df, 7)
str(df)
summary(df)
A data.frame: 7 × 15
Country.RegionConfirmedDeathsRecoveredActiveNew.casesNew.deathsNew.recoveredDeaths...100.CasesRecovered...100.CasesDeaths...100.RecoveredConfirmed.last.weekX1.week.changeX1.week...increaseWHO.Region
<chr><int><int><int><int><int><int><int><dbl><dbl><dbl><int><int><dbl><chr>
1Afghanistan 36263126925198 9796 106 10 183.5069.49 5.04 35526 737 2.07Eastern Mediterranean
2Albania 4880 144 2745 1991 117 6 632.9556.25 5.25 4171 70917.00Europe
3Algeria 27973116318837 7973 616 8 7494.1667.34 6.17 23691 428218.07Africa
4Andorra 907 52 803 52 10 0 05.7388.53 6.48 884 23 2.60Europe
5Angola 950 41 242 667 18 1 04.3225.4716.94 749 20126.84Africa
6Antigua and Barbuda 86 3 65 18 4 0 53.4975.58 4.62 76 1013.16Americas
7Argentina 16741630597257591782489012020571.8343.35 4.211307743664228.02Americas 
'data.frame':	187 obs. of  15 variables:
 $ Country.Region        : chr  "Afghanistan" "Albania" "Algeria" "Andorra" ...
 $ Confirmed             : int  36263 4880 27973 907 950 86 167416 37390 15303 20558 ...
 $ Deaths                : int  1269 144 1163 52 41 3 3059 711 167 713 ...
 $ Recovered             : int  25198 2745 18837 803 242 65 72575 26665 9311 18246 ...
 $ Active                : int  9796 1991 7973 52 667 18 91782 10014 5825 1599 ...
 $ New.cases             : int  106 117 616 10 18 4 4890 73 368 86 ...
 $ New.deaths            : int  10 6 8 0 1 0 120 6 6 1 ...
 $ New.recovered         : int  18 63 749 0 0 5 2057 187 137 37 ...
 $ Deaths...100.Cases    : num  3.5 2.95 4.16 5.73 4.32 3.49 1.83 1.9 1.09 3.47 ...
 $ Recovered...100.Cases : num  69.5 56.2 67.3 88.5 25.5 ...
 $ Deaths...100.Recovered: num  5.04 5.25 6.17 6.48 16.94 ...
 $ Confirmed.last.week   : int  35526 4171 23691 884 749 76 130774 34981 12428 19743 ...
 $ X1.week.change        : int  737 709 4282 23 201 10 36642 2409 2875 815 ...
 $ X1.week...increase    : num  2.07 17 18.07 2.6 26.84 ...
 $ WHO.Region            : chr  "Eastern Mediterranean" "Europe" "Africa" "Europe" ...

 Country.Region       Confirmed           Deaths           Recovered        
 Length:187         Min.   :     10   Min.   :     0.0   Min.   :      0.0  
 Class :character   1st Qu.:   1114   1st Qu.:    18.5   1st Qu.:    626.5  
 Mode  :character   Median :   5059   Median :   108.0   Median :   2815.0  
                    Mean   :  88131   Mean   :  3497.5   Mean   :  50631.5  
                    3rd Qu.:  40460   3rd Qu.:   734.0   3rd Qu.:  22606.0  
                    Max.   :4290259   Max.   :148011.0   Max.   :1846641.0  
     Active            New.cases         New.deaths      New.recovered    
 Min.   :      0.0   Min.   :    0.0   Min.   :   0.00   Min.   :    0.0  
 1st Qu.:    141.5   1st Qu.:    4.0   1st Qu.:   0.00   1st Qu.:    0.0  
 Median :   1600.0   Median :   49.0   Median :   1.00   Median :   22.0  
 Mean   :  34001.9   Mean   : 1223.0   Mean   :  28.96   Mean   :  933.8  
 3rd Qu.:   9149.0   3rd Qu.:  419.5   3rd Qu.:   6.00   3rd Qu.:  221.0  
 Max.   :2816444.0   Max.   :56336.0   Max.   :1076.00   Max.   :33728.0  
 Deaths...100.Cases Recovered...100.Cases Deaths...100.Recovered
 Min.   : 0.000     Min.   :  0.00        Min.   :0.00          
 1st Qu.: 0.945     1st Qu.: 48.77        1st Qu.:1.45          
 Median : 2.150     Median : 71.32        Median :3.62          
 Mean   : 3.020     Mean   : 64.82        Mean   : Inf          
 3rd Qu.: 3.875     3rd Qu.: 86.89        3rd Qu.:6.44          
 Max.   :28.560     Max.   :100.00        Max.   : Inf          
 Confirmed.last.week X1.week.change   X1.week...increase  WHO.Region       
 Min.   :     10     Min.   :   -47   Min.   : -3.840    Length:187        
 1st Qu.:   1052     1st Qu.:    49   1st Qu.:  2.775    Class :character  
 Median :   5020     Median :   432   Median :  6.890    Mode  :character  
 Mean   :  78682     Mean   :  9448   Mean   : 13.606                      
 3rd Qu.:  37080     3rd Qu.:  3172   3rd Qu.: 16.855                      
 Max.   :3834677     Max.   :455582   Max.   :226.320
  • Are there any missing data?
In [4]:
vis_miss(df)

Warning message:
"`gather_()` was deprecated in tidyr 1.2.0.
ℹ Please use `gather()` instead.
ℹ The deprecated feature was likely used in the visdat package.
  Please report the issue at <https://github.com/ropensci/visdat/issues>."
In [5]:
# Removed a row with a strange value
df = df[!(df$Deaths...100.Recovered=="Inf" ),]
  • How each variable is distributed?
  • Are there any strange behaviour?

Let's explore the dataset

EDA¶

In [6]:
print(colnames(df))

 [1] "Country.Region"         "Confirmed"              "Deaths"                
 [4] "Recovered"              "Active"                 "New.cases"             
 [7] "New.deaths"             "New.recovered"          "Deaths...100.Cases"    
[10] "Recovered...100.Cases"  "Deaths...100.Recovered" "Confirmed.last.week"   
[13] "X1.week.change"         "X1.week...increase"     "WHO.Region"

Univariate EDA¶

Confirmed¶

In [7]:
boxplot.stats(df$Confirmed)
$stats
  1. 10
  2. 1100
  3. 4970
  4. 39741
  5. 92482
$n
182
$conf
  1. 444.466854473945
  2. 9495.53314552605
$out
  1. 167416
  2. 226225
  3. 2442375
  4. 347923
  5. 257101
  6. 220352
  7. 207112
  8. 1480073
  9. 100303
  10. 293606
  11. 112585
  12. 246286
  13. 395489
  14. 274289
  15. 389717
  16. 109597
  17. 816680
  18. 268934
  19. 452529
  20. 272421
  21. 227019
  22. 4290259
  23. 301708
In [8]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Confirmed)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of confirmed cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Confirmed)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of confirmed cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000))
In [9]:
ggplot(df, aes(x=Confirmed)) +
    geom_histogram( binwidth=4000, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of confirmed cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,120000)

Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [10]:
png(filename = "img/Confirmed_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Confirmed)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of confirmed cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Confirmed_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Confirmed)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of confirmed cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000))

dev.off()

png(filename = "img/Confirmed_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Confirmed)) +
    geom_histogram( binwidth=4000, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of confirmed cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,120000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Deaths¶

In [11]:
boxplot.stats(df$Deaths)
$stats
  1. 0
  2. 18
  3. 103.5
  4. 720
  5. 1764
$n
182
$conf
  1. 21.2835907932173
  2. 185.716409206783
$out
  1. 3059
  2. 2965
  3. 9822
  4. 2647
  5. 87618
  6. 9187
  7. 4656
  8. 8777
  9. 5532
  10. 4652
  11. 30212
  12. 9125
  13. 33408
  14. 4838
  15. 15912
  16. 4458
  17. 35112
  18. 44022
  19. 6160
  20. 5842
  21. 18418
  22. 1945
  23. 2206
  24. 13334
  25. 2760
  26. 7067
  27. 28432
  28. 1978
  29. 5630
  30. 148011
  31. 45844
In [12]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 5000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of deaths cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 5000))
In [13]:
ggplot(df, aes(Deaths)) +
    geom_histogram( binwidth=150, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of deaths cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,5000)

Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [14]:
png(filename = "img/Deaths_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 5000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Deaths_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of deaths cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 5000))

dev.off()

png(filename = "img/Deaths_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Deaths)) +
    geom_histogram( binwidth=150, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of deaths cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,5000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Recovered¶

In [15]:
boxplot.stats(df$Recovered)
$stats
  1. 0
  2. 714
  3. 3220
  4. 23364
  5. 57028
$n
182
$conf
  1. 567.291070464917
  2. 5872.70892953508
$out
  1. 72575
  2. 125683
  3. 60492
  4. 1846641
  5. 319954
  6. 78869
  7. 131161
  8. 81212
  9. 190314
  10. 951166
  11. 58173
  12. 255144
  13. 77144
  14. 198593
  15. 303810
  16. 241026
  17. 272547
  18. 106328
  19. 602249
  20. 222936
  21. 274925
  22. 150376
  23. 210469
  24. 1325804
In [16]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of recovered cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 70000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of recovered cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 70000))
In [17]:
ggplot(df, aes(x=Recovered)) +
    geom_histogram( binwidth=3000, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of recovered cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,70000)

Warning message:
"Removed 22 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [18]:
png(filename = "img/Recovered_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of recovered cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 70000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Recovered_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of confirmed cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 70000))

dev.off()

png(filename = "img/Recovered_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Recovered)) +
    geom_histogram( binwidth=3000, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of confirmed cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,70000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 22 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Active¶

In [19]:
boxplot.stats(df$Active)
$stats
  1. 0
  2. 129
  3. 1599
  4. 8870
  5. 19637
$n
182
$conf
  1. 575.27687624432
  2. 2622.72312375568
$out
  1. 91782
  2. 97577
  3. 39154
  4. 47056
  5. 508116
  6. 117163
  7. 32869
  8. 40733
  9. 52992
  10. 108928
  11. 33536
  12. 495499
  13. 37292
  14. 22550
  15. 30983
  16. 36378
  17. 29659
  18. 47657
  19. 47064
  20. 22117
  21. 27421
  22. 25034
  23. 98752
  24. 53649
  25. 201097
  26. 43238
  27. 170537
  28. 93613
  29. 2816444
  30. 28258
  31. 254427
In [20]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Active)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of active cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 40000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Active)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of active cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 40000))
In [21]:
ggplot(df, aes(x=Active)) +
    geom_histogram( binwidth=2500, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of active cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,40000)

Warning message:
"Removed 19 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [22]:
png(filename = "img/Active_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Active)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of active cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 40000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Active_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Active)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of active cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 40000))

dev.off()

png(filename = "img/Active_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Active)) +
    geom_histogram( binwidth=2500, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of active cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,40000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 19 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

New cases¶

In [23]:
boxplot.stats(df$New.cases)
$stats
  1. 0
  2. 4
  3. 49
  4. 420
  5. 919
$n
182
$conf
  1. 0.279164914499141
  2. 97.7208350855009
$out
  1. 4890
  2. 2772
  3. 1752
  4. 23284
  5. 2133
  6. 16306
  7. 1248
  8. 2551
  9. 44457
  10. 1525
  11. 2434
  12. 2553
  13. 2029
  14. 1526
  15. 4973
  16. 1053
  17. 1176
  18. 1146
  19. 13756
  20. 1592
  21. 1104
  22. 5607
  23. 1993
  24. 7096
  25. 56336
In [24]:
ggplot(df, aes(x=as.factor(WHO.Region), y=New.cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1600)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=New.cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1600))
In [25]:
ggplot(df, aes(x=New.cases)) +
    geom_histogram( binwidth=70, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of new cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,1600)

Warning message:
"Removed 17 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [26]:
png(filename = "img/New_cases_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=New.cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1600)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/New_cases_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=New.cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1600))

dev.off()

png(filename = "img/New_cases_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(New.cases)) +
    geom_histogram( binwidth=70, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of new cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,1600)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 17 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

New deaths¶

In [27]:
boxplot.stats(df$New.deaths)
$stats
  1. 0
  2. 0
  3. 1
  4. 6
  5. 14
$n
182
$conf
  1. 0.297295647805276
  2. 1.70270435219472
$out
  1. 120
  2. 37
  3. 64
  4. 614
  5. 75
  6. 508
  7. 20
  8. 17
  9. 46
  10. 17
  11. 27
  12. 50
  13. 637
  14. 57
  15. 212
  16. 96
  17. 16
  18. 24
  19. 342
  20. 20
  21. 28
  22. 575
  23. 19
  24. 85
  25. 27
  26. 298
  27. 17
  28. 1076
In [28]:
ggplot(df, aes(x=as.factor(WHO.Region), y=New.deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new deaths cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 60)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=New.deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new deaths cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 60))
In [29]:
ggplot(df, aes(x=New.deaths)) +
    geom_histogram( binwidth=5, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of new deaths cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,60)

Warning message:
"Removed 13 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [30]:
png(filename = "img/New_deaths_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=New.deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new deaths cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 60)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/New_deaths_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=New.deaths)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new deaths cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 60))

dev.off()

png(filename = "img/New_deaths_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(New.deaths)) +
    geom_histogram( binwidth=5, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of new deaths cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,60)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 13 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

New recovered¶

In [31]:
boxplot.stats(df$New.recovered)
$stats
  1. 0
  2. 0
  3. 24
  4. 230
  5. 569
$n
182
$conf
  1. -2.93700016746442
  2. 50.9370001674644
$out
  1. 749
  2. 2057
  3. 1801
  4. 33728
  5. 1859
  6. 11494
  7. 1601
  8. 1007
  9. 843
  10. 33598
  11. 1518
  12. 1931
  13. 1927
  14. 1833
  15. 684
  16. 817
  17. 681
  18. 8588
  19. 626
  20. 829
  21. 1729
  22. 3592
  23. 955
  24. 4697
  25. 3077
  26. 2613
  27. 9848
  28. 982
  29. 27941
In [32]:
ggplot(df, aes(x=as.factor(WHO.Region), y=New.recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new recovered cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1700)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=New.recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new recovered cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1700))
In [33]:
ggplot(df, aes(x=New.recovered)) +
    geom_histogram( binwidth=80, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of new recovered cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,1700)

Warning message:
"Removed 17 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [34]:
png(filename = "img/New_recovered_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=New.recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of new recovered cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1700)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/New_recovered_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=New.recovered)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of new recovered cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 1700))

dev.off()

png(filename = "img/New_recovered_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(New.recovered)) +
    geom_histogram( binwidth=80, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of new recovered cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,1700)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 17 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Deaths / 100 Cases¶

In [35]:
boxplot.stats(df$Deaths...100.Cases)
$stats
  1. 0
  2. 0.94
  3. 2.13
  4. 3.72
  5. 6.82
$n
182
$conf
  1. 1.80441365014978
  2. 2.45558634985022
$out
  1. 14.79
  2. 8.13
  3. 13.71
  4. 13.4
  5. 14.26
  6. 11.13
  7. 11.53
  8. 10.44
  9. 15.19
  10. 10
  11. 28.56
In [36]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 10)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of deaths over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 10))
In [37]:
ggplot(df, aes(x=Deaths...100.Cases)) +
    geom_histogram( binwidth=0.6, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of deaths over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,10)

Warning message:
"Removed 9 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [38]:
png(filename = "img/Deaths_over_100_cases_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 10)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Deaths_over_100_cases_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of deaths over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 10))

dev.off()

png(filename = "img/Deaths_over_100_cases_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Deaths...100.Cases)) +
    geom_histogram( binwidth=0.6, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of deaths over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,10)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 9 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Recovered / 100 Cases¶

In [39]:
boxplot.stats(df$Recovered...100.Cases)
$stats
  1. 0
  2. 51.73
  3. 72.27
  4. 87.41
  5. 100
$n
182
$conf
  1. 68.091251452282
  2. 76.448748547718
$out
In [40]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Recovered...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of recovered over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 110)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Recovered...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of recovered over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 110))
In [41]:
ggplot(df, aes(x=Recovered...100.Cases)) +
    geom_histogram( binwidth=8, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of recovered over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,110)

Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [42]:
png(filename = "img/Recovered_over_100_cases_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Recovered...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of recovered over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 110)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Recovered_over_100_cases_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Confirmed)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of recovered over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 110))

dev.off()

png(filename = "img/Recovered_over_100_cases_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Recovered...100.Cases)) +
    geom_histogram( binwidth=8, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of recovered over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,110)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Deaths / 100 Recovered¶

In [43]:
boxplot.stats(df$Deaths...100.Cases)
$stats
  1. 0
  2. 0.94
  3. 2.13
  4. 3.72
  5. 6.82
$n
182
$conf
  1. 1.80441365014978
  2. 2.45558634985022
$out
  1. 14.79
  2. 8.13
  3. 13.71
  4. 13.4
  5. 14.26
  6. 11.13
  7. 11.53
  8. 10.44
  9. 15.19
  10. 10
  11. 28.56
In [44]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 15)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of recovered over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 15))
In [45]:
ggplot(df, aes(x=Deaths...100.Cases)) +
    geom_histogram( binwidth=0.8, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of deaths over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,15)

Warning message:
"Removed 2 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [46]:
png(filename = "img/Deaths_over_100_cases_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of deaths over 100 cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 15)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Deaths_over_100_cases_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Deaths...100.Cases)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of deaths over 100 cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 15))

dev.off()

png(filename = "img/Deaths_over_100_cases_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Deaths...100.Cases)) +
    geom_histogram( binwidth=0.8, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of deaths over 100 cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,15)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 2 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

Confirmed last week¶

In [47]:
boxplot.stats(df$Confirmed.last.week)
$stats
  1. 10
  2. 1039
  3. 4784
  4. 36936
  5. 88402
$n
182
$conf
  1. 579.836978210999
  2. 8988.163021789
$out
  1. 130774
  2. 207453
  3. 2118646
  4. 333029
  5. 204005
  6. 214023
  7. 203325
  8. 1155338
  9. 276202
  10. 94693
  11. 244624
  12. 349396
  13. 266096
  14. 357681
  15. 107037
  16. 776212
  17. 253349
  18. 373628
  19. 264836
  20. 220572
  21. 3834677
  22. 296944
In [48]:
ggplot(df, aes(x=as.factor(WHO.Region), y=Confirmed.last.week)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of confirmed last week cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=Confirmed.last.week)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of confirmed last week cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000))
In [49]:
ggplot(df, aes(x=Confirmed.last.week)) +
    geom_histogram( binwidth=5000, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of confirmed last week cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,120000)

Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [50]:
png(filename = "img/Confirmed_last_week_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=Confirmed.last.week)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of confirmed last week cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/Confirmed_last_week_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=Confirmed.last.week)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of confirmed last week cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 120000))

dev.off()

png(filename = "img/Confirmed_last_week_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Confirmed.last.week)) +
    geom_histogram( binwidth=5000, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of confirmed last week cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,120000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 20 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

1 week change¶

In [51]:
boxplot.stats(df$X1.week.change)
$stats
  1. -47
  2. 47
  3. 418
  4. 3325
  5. 8193
$n
182
$conf
  1. 34.0891889176158
  2. 801.910811082384
$out
  1. 36642
  2. 18772
  3. 10190
  4. 323729
  5. 14894
  6. 53096
  7. 10200
  8. 324735
  9. 12089
  10. 17404
  11. 17892
  12. 11982
  13. 11180
  14. 46093
  15. 8658
  16. 32036
  17. 13142
  18. 40468
  19. 15585
  20. 78901
  21. 455582
In [52]:
ggplot(df, aes(x=as.factor(WHO.Region), y=X1.week.change)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of 1 week change cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 11000)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=X1.week.change)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of 1 week change cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 11000))
In [53]:
ggplot(df, aes(x=X1.week.change)) +
    geom_histogram( binwidth=500, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of 1 week change cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,11000)

Warning message:
"Removed 19 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [54]:
png(filename = "img/1_week_change_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=X1.week.change)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of 1 week change cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 11000)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/1_week_change_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=X1.week.change)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of 1 week change cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 11000))

dev.off()

png(filename = "img/1_week_change_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(X1.week.change)) +
    geom_histogram( binwidth=500, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of 1 week change cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,11000)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 19 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

1 week % increase¶

In [55]:
boxplot.stats(df$X1.week...increase)
$stats
  1. -3.84
  2. 2.75
  3. 6.89
  4. 17
  5. 37.44
$n
182
$conf
  1. 5.22107716353753
  2. 8.55892283646247
$out
  1. 119.54
  2. 41.57
  3. 42.52
  4. 191.07
  5. 40.67
  6. 42.78
  7. 226.32
  8. 57.85
In [56]:
ggplot(df, aes(x=as.factor(WHO.Region), y=X1.week...increase)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of 1 week increase cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 50)) + 
    xlab("WHO.Region")

ggplot(df, aes(y=X1.week...increase)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of 1 week increase cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 50))
In [57]:
ggplot(df, aes(x=X1.week...increase)) +
    geom_histogram( binwidth=3, fill="#69b3a2", color="#e9ecef", alpha=0.9, position = 'identity') +
    ggtitle("Histogram of 1 week increase cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,50)

Warning message:
"Removed 5 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
In [58]:
png(filename = "img/1_week_increase_conditional_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=as.factor(WHO.Region), y=X1.week...increase)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplots of 1 week increase cases divided by WHO region") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14,face="bold")
    ) +
    coord_cartesian(ylim = c(0, 50)) +
    xlab("WHO.Region")

dev.off()

png(filename = "img/1_week_increase_boxplot.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(y=X1.week...increase)) + 
    geom_boxplot(fill="slateblue", alpha=0.2) + 
    ggtitle("Boxplot of 1 week increase cases") +
    theme(
        axis.text=element_text(size=6.5), 
        axis.title=element_text(size=14, face="bold")
    ) +
    coord_cartesian(ylim = c(0, 50))

dev.off()

png(filename = "img/1_week_increase_histogram.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(Confirmed)) +
    geom_histogram( binwidth=3, fill="#69b3a2", color="#e9ecef", alpha=0.9) +
    ggtitle("Histogram of 1 week increase cases") +
    theme(plot.title = element_text(size=15)) +
    xlim(0,50)

dev.off()
png: 2
png: 2 
Warning message:
"Removed 171 rows containing non-finite values (`stat_bin()`)."
Warning message:
"Removed 2 rows containing missing values (`geom_bar()`)."
png: 2

WHO Region¶

In [59]:
table(df$WHO.Region)

               Africa              Americas Eastern Mediterranean 
                   47                    34                    21 
               Europe       South-East Asia       Western Pacific 
                   54                    10                    16

Multivariate EDA¶

  • How does the variables jointly behave?
  • Is there a correlation between confirmed cases, deaths cases and recovered cases?
In [60]:
png(filename = "img/Scatter_confirmed_vs_deaths.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=Confirmed, y=Deaths, color=WHO.Region)) + 
    geom_point(size=3) +
    geom_smooth(method=lm , color="red", fill="#69b3a2", se=TRUE) +
    xlim(0,100000) +
    ylim(0,4000) +
    theme(legend.position = c(0.8, 0.8))

dev.off()


ggplot(df, aes(x=Confirmed, y=Deaths, color=WHO.Region)) + 
    geom_point(size=3) +
    geom_smooth(method=lm , color="red", fill="#69b3a2", se=TRUE) +
    xlim(0,100000) +
    ylim(0,4000) +
    theme(legend.position = c(0.85, 0.85))

`geom_smooth()` using formula = 'y ~ x'
Warning message:
"Removed 28 rows containing non-finite values (`stat_smooth()`)."
Warning message:
"Removed 28 rows containing missing values (`geom_point()`)."
png: 2 
`geom_smooth()` using formula = 'y ~ x'
Warning message:
"Removed 28 rows containing non-finite values (`stat_smooth()`)."
Warning message:
"Removed 28 rows containing missing values (`geom_point()`)."
In [61]:
confirmed_deaths_lm = lm(formula = Deaths ~ Confirmed, data=df)
summary(confirmed_deaths_lm)

Call:
lm(formula = Deaths ~ Confirmed, data = df)

Residuals:
   Min     1Q Median     3Q    Max 
-17907   -574   -456   -435  35034 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) 4.393e+02  3.864e+02   1.137    0.257    
Confirmed   3.437e-02  9.720e-04  35.363   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 5080 on 180 degrees of freedom
Multiple R-squared:  0.8742,	Adjusted R-squared:  0.8735 
F-statistic:  1251 on 1 and 180 DF,  p-value: < 2.2e-16
In [62]:
png(filename = "img/Scatter_confirmed_vs_recovered.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=Confirmed, y=Recovered, color=WHO.Region)) + 
    geom_point(size=3) +
    geom_smooth(method=lm , color="red", fill="#69b3a2", se=TRUE) +
    xlim(0,100000) +
    ylim(0,100000) +
    theme(legend.position = c(0.8, 0.8))

dev.off()


ggplot(df, aes(x=Confirmed, y=Recovered, color=WHO.Region)) + 
    geom_point(size=3) +
    geom_smooth(method=lm , color="red", fill="#69b3a2", se=TRUE) +
    xlim(0,100000) +
    ylim(0,100000) +
    theme(legend.position = c(0.85, 0.85))

`geom_smooth()` using formula = 'y ~ x'
Warning message:
"Removed 23 rows containing non-finite values (`stat_smooth()`)."
Warning message:
"Removed 23 rows containing missing values (`geom_point()`)."
png: 2 
`geom_smooth()` using formula = 'y ~ x'
Warning message:
"Removed 23 rows containing non-finite values (`stat_smooth()`)."
Warning message:
"Removed 23 rows containing missing values (`geom_point()`)."
In [63]:
confirmed_recovered_lm = lm(formula = Recovered ~ Confirmed, data=df)
summary(confirmed_recovered_lm)

Call:
lm(formula = Recovered ~ Confirmed, data = df)

Residuals:
    Min      1Q  Median      3Q     Max 
-615078  -11847  -11340   -5746  736621 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) 1.186e+04  6.195e+03   1.914   0.0572 .  
Confirmed   4.496e-01  1.558e-02  28.853   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 81440 on 180 degrees of freedom
Multiple R-squared:  0.8222,	Adjusted R-squared:  0.8212 
F-statistic: 832.5 on 1 and 180 DF,  p-value: < 2.2e-16
In [64]:
df_num = df[2:14]

# First
corrgram(df_num, order=TRUE, lower.panel=panel.shade, upper.panel=panel.pie, text.panel=panel.txt, main="Correlation matrix") 

png(filename = "img/Correlation_matrix.png", units="in", width=5, height=5, res=300)

corrgram(df_num, order=TRUE, lower.panel=panel.shade, upper.panel=panel.pie, text.panel=panel.txt, main="Correlation matrix") 

dev.off()
png: 2

Dimensionality reduction¶

  • Can the dataset be resumed by in a lower dimensional space?
  • If yes, is this rappresention useful to highlight some behaviour in the data?

Let's try PCA!

In [65]:
str(df)

'data.frame':	182 obs. of  15 variables:
 $ Country.Region        : chr  "Afghanistan" "Albania" "Algeria" "Andorra" ...
 $ Confirmed             : int  36263 4880 27973 907 950 86 167416 37390 15303 20558 ...
 $ Deaths                : int  1269 144 1163 52 41 3 3059 711 167 713 ...
 $ Recovered             : int  25198 2745 18837 803 242 65 72575 26665 9311 18246 ...
 $ Active                : int  9796 1991 7973 52 667 18 91782 10014 5825 1599 ...
 $ New.cases             : int  106 117 616 10 18 4 4890 73 368 86 ...
 $ New.deaths            : int  10 6 8 0 1 0 120 6 6 1 ...
 $ New.recovered         : int  18 63 749 0 0 5 2057 187 137 37 ...
 $ Deaths...100.Cases    : num  3.5 2.95 4.16 5.73 4.32 3.49 1.83 1.9 1.09 3.47 ...
 $ Recovered...100.Cases : num  69.5 56.2 67.3 88.5 25.5 ...
 $ Deaths...100.Recovered: num  5.04 5.25 6.17 6.48 16.94 ...
 $ Confirmed.last.week   : int  35526 4171 23691 884 749 76 130774 34981 12428 19743 ...
 $ X1.week.change        : int  737 709 4282 23 201 10 36642 2409 2875 815 ...
 $ X1.week...increase    : num  2.07 17 18.07 2.6 26.84 ...
 $ WHO.Region            : chr  "Eastern Mediterranean" "Europe" "Africa" "Europe" ...
In [66]:
df_num = df[2:14]

df_scaled = scale(df_num)

corr_mat = cor(df_scaled)
eig = eigen(corr_mat)

PVE = eig$values/sum(eig$values)
PVE

cumsum(PVE)
  1. 0.616977459780826
  2. 0.123266129637276
  3. 0.104685583846847
  4. 0.0510483067850515
  5. 0.0379863874959457
  6. 0.0319526100795748
  7. 0.0199390336020361
  8. 0.00887849217753967
  9. 0.00361606840007441
  10. 0.00130097169141562
  11. 0.000348956503413679
  12. -7.63943361031493e-18
  13. -2.45989981986085e-17
  1. 0.616977459780826
  2. 0.740243589418102
  3. 0.844929173264948
  4. 0.89597748005
  5. 0.933963867545946
  6. 0.965916477625521
  7. 0.985855511227557
  8. 0.994734003405096
  9. 0.998350071805171
  10. 0.999651043496586
  11. 1
  12. 1
  13. 1
In [67]:
df_pca = prcomp(df_scaled)

# compute total variance
variance = df_pca$sdev^2 / sum(df_pca$sdev^2)

# Scree plot
qplot(c(1:13), variance) +
  geom_line() +
  geom_point(size=4)+
  xlab("Principal Component") +
  ylab("Variance Explained") +
  ggtitle("Scree Plot") +
  ylim(0, 1)

Warning message:
"`qplot()` was deprecated in ggplot2 3.4.0."
In [68]:
# compute total variance
cum_variance = cumsum(variance)

# Scree plot
qplot(c(1:13), cum_variance) +
  geom_line() +
  geom_point(size=4)+
  xlab("Principal Component") +
  ylab("Variance Explained") +
  ggtitle("Scree Plot") +
  ylim(0, 1)
In [69]:
# Kaiser rule
kaiser = df_pca$sdev^2
print(kaiser)

which(kaiser>1)

 [1] 8.020707e+00 1.602460e+00 1.360913e+00 6.636280e-01 4.938230e-01
 [6] 4.153839e-01 2.592074e-01 1.154204e-01 4.700889e-02 1.691263e-02
[11] 4.536435e-03 2.061963e-31 1.867706e-33
  1. 1
  2. 2
  3. 3
In [70]:
#str(df_pca)

pca_scores = data.frame(df_pca$x[,1:2])
plot(pca_scores)

biplot(df_pca)

png(filename = "img/Biplot.png", units="in", width=5, height=5, res=300)

biplot(df_pca)

dev.off()
png: 2

Predictive Analysis¶

Clustering¶

  • Are there any hidden substructure in the data?
  • If there's any, are they correspondent with WHO.Region?

Let's try to clusterize the data!

In [71]:
k = 6
fit_kmeans <- kmeans(df_scaled, k)
In [72]:
table(df$WHO.Region, fit_kmeans$cluster)

                       
                         1  2  3  4  5  6
  Africa                 1 23  0 22  1  0
  Americas               1 14  2 14  3  0
  Eastern Mediterranean  0 13  0  6  2  0
  Europe                 0 33  0 13  6  2
  South-East Asia        0  6  1  3  0  0
  Western Pacific        1 12  0  3  0  0
In [80]:
ggplot(df, aes(x=Confirmed, y=Deaths, color=factor(fit_kmeans$cluster))) + 
    geom_point(size=3) +
    xlim(0,100000) +
    ylim(0,4000) +
    theme(legend.position = c(0.85, 0.85))

png(filename = "img/Scatter_confirmed_vs_deaths_cluster.png", units="in", width=5, height=5, res=300)

ggplot(df, aes(x=Confirmed, y=Deaths, color=factor(fit_kmeans$cluster))) + 
    geom_point(size=3) +
    xlim(0,100000) +
    ylim(0,4000) +
    theme(legend.position = c(0.85, 0.85))

dev.off()

Warning message:
"Removed 28 rows containing missing values (`geom_point()`)."
Warning message:
"Removed 28 rows containing missing values (`geom_point()`)."
png: 2
In [81]:
ggplot(df, aes(x=Confirmed, y=Deaths, color=WHO.Region)) + 
    geom_point(size=3) +
    xlim(0,100000) +
    ylim(0,4000) +
    theme(legend.position = c(0.85, 0.85))

Warning message:
"Removed 28 rows containing missing values (`geom_point()`)."
In [75]:
pca_dataframe = data.frame(df_pca$x)
In [76]:
ggplot(pca_dataframe, aes(x=PC1, y=PC2, color=factor(fit_kmeans$cluster))) + 
    geom_point(size=3) +
    theme(legend.position = c(0.85, 0.85))

png(filename = "img/biplot_cluster.png", units="in", width=5, height=5, res=300)

ggplot(pca_dataframe, aes(x=PC1, y=PC2, color=factor(fit_kmeans$cluster))) + 
    geom_point(size=3) +
    theme(legend.position = c(0.85, 0.85))

dev.off()
png: 2
In [77]:
ggplot(pca_dataframe, aes(x=PC1, y=PC2, color=df$WHO.Region)) + 
    geom_point(size=3) +
    theme(legend.position = c(0.85, 0.85))

png(filename = "img/biplot_whoregion.png", units="in", width=5, height=5, res=300)

ggplot(pca_dataframe, aes(x=PC1, y=PC2, color=df$WHO.Region)) + 
    geom_point(size=3) +
    theme(legend.position = c(0.85, 0.85))

dev.off()
png: 2

Linear regression¶

  • Can we predict the deaths cases on the other values?

Let's try a linear regressor

In [78]:
linear_model = lm(formula = Deaths ~ Confirmed + Recovered + Active + New.cases + New.deaths + 
                      New.recovered + Deaths...100.Cases + Recovered...100.Cases + 
                      Deaths...100.Recovered + Confirmed.last.week + X1.week.change +
                      X1.week...increase, data=df)
In [79]:
summary(linear_model)

Call:
lm(formula = Deaths ~ Confirmed + Recovered + Active + New.cases + 
    New.deaths + New.recovered + Deaths...100.Cases + Recovered...100.Cases + 
    Deaths...100.Recovered + Confirmed.last.week + X1.week.change + 
    X1.week...increase, data = df)

Residuals:
       Min         1Q     Median         3Q        Max 
-3.938e-10 -1.841e-11 -1.111e-11 -5.690e-12  1.754e-09 

Coefficients: (1 not defined because of singularities)
                         Estimate Std. Error    t value Pr(>|t|)    
(Intercept)             3.452e-11  4.622e-11  7.470e-01    0.456    
Confirmed               1.000e+00  4.228e-15  2.365e+14   <2e-16 ***
Recovered              -1.000e+00  3.073e-15 -3.254e+14   <2e-16 ***
Active                 -1.000e+00  3.133e-15 -3.191e+14   <2e-16 ***
New.cases              -5.663e-15  1.179e-14 -4.800e-01    0.632    
New.deaths             -4.424e-14  3.589e-13 -1.230e-01    0.902    
New.recovered           1.051e-15  2.351e-14  4.500e-02    0.964    
Deaths...100.Cases      7.179e-13  3.990e-12  1.800e-01    0.857    
Recovered...100.Cases  -1.770e-13  5.665e-13 -3.120e-01    0.755    
Deaths...100.Recovered -5.731e-15  3.957e-14 -1.450e-01    0.885    
Confirmed.last.week    -9.233e-16  2.783e-15 -3.320e-01    0.740    
X1.week.change                 NA         NA         NA       NA    
X1.week...increase      5.668e-14  5.230e-13  1.080e-01    0.914    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 1.502e-10 on 170 degrees of freedom
Multiple R-squared:      1,	Adjusted R-squared:      1 
F-statistic: 1.486e+29 on 11 and 170 DF,  p-value: < 2.2e-16
In [ ]: