Introduction

We import the data from the official page of the dataset.

Example 1.1: Munich Rent Index

munich_rent_url <- "https://www.uni-goettingen.de/de/document/download/64c29c1b1fccb142cfa8f29a942a9e05.raw/rent99.raw"

munich_rent_index <- read.table(
  url(munich_rent_url),
  header = 1,
  colClasses = c(
    "numeric", "numeric", "numeric",
    "numeric", "factor", "factor",
    "factor", "factor", "factor"
  )
)

# Convert to logical
munich_rent_index$bath <- munich_rent_index$bath == 1
munich_rent_index$kitchen <- munich_rent_index$kitchen == 1
munich_rent_index$cheating <- munich_rent_index$cheating == 1
summary(munich_rent_index)

      rent            rentsqm             area            yearc      location
 Min.   :  40.51   Min.   : 0.4158   Min.   : 20.00   Min.   :1918   1:1794  
 1st Qu.: 322.03   1st Qu.: 5.2610   1st Qu.: 51.00   1st Qu.:1939   2:1210  
 Median : 426.97   Median : 6.9802   Median : 65.00   Median :1959   3:  78  
 Mean   : 459.44   Mean   : 7.1113   Mean   : 67.37   Mean   :1956           
 3rd Qu.: 559.36   3rd Qu.: 8.8408   3rd Qu.: 81.00   3rd Qu.:1972           
 Max.   :1843.38   Max.   :17.7216   Max.   :160.00   Max.   :1997           
                                                                             
    bath          kitchen         cheating          district   
 Mode :logical   Mode :logical   Mode :logical   411    :  53  
 FALSE:2891      FALSE:2951      FALSE:321       623    :  53  
 TRUE :191       TRUE :131       TRUE :2761      350    :  49  
                                                 563    :  49  
                                                 711    :  42  
                                                 360    :  38  
                                                 (Other):2798  

For future use, this code is going to be available in an R script in import_data/munich_rent_index.R.

Example 2.2: Malnutrition in Zambia

zambia_url <- "https://www.uni-goettingen.de/de/document/download/d90a2d7b26c4504ab6630cf36cbae2fa.raw/zambia_height92.raw"

malnutrition_zambia <- read.table(
  url(zambia_url),
  header = 1,
  colClasses = c(
    "numeric", "factor", "numeric",
    "integer", "numeric", "numeric",
    "numeric", "factor", "factor",
    "factor", "factor", "factor"
  )
)

# Convert to logical
malnutrition_zambia$m_work <- malnutrition_zambia$m_work == 1
summary(malnutrition_zambia)

     zscore       c_gender   c_breastf         c_age         m_agebirth   
 Min.   :-600.0   0:2254   Min.   : 0.00   Min.   : 0.00   Min.   :13.17  
 1st Qu.:-257.0   1:2167   1st Qu.: 1.00   1st Qu.:12.00   1st Qu.:21.08  
 Median :-171.0            Median :14.00   Median :26.00   Median :25.33  
 Mean   :-171.2            Mean   :11.12   Mean   :27.12   Mean   :26.40  
 3rd Qu.: -86.0            3rd Qu.:19.00   3rd Qu.:42.00   3rd Qu.:31.08  
 Max.   : 503.0            Max.   :46.00   Max.   :59.00   Max.   :48.67  
                                                                          
    m_height         m_bmi       m_education   m_work           district   
 Min.   :134.0   Min.   :13.15   1: 822      Mode :logical   98     : 488  
 1st Qu.:154.0   1st Qu.:19.75   2:2756      FALSE:1978      61     : 270  
 Median :158.1   Median :21.46   3: 767      TRUE :2443      53     : 235  
 Mean   :158.1   Mean   :22.00   4:  76                      97     : 156  
 3rd Qu.:162.0   3rd Qu.:23.57                               94     : 153  
 Max.   :185.0   Max.   :39.29                               66     : 128  
                                                             (Other):2991  
     region    time    
 2      :967   1:4421  
 8      :659           
 5      :609           
 6      :430           
 3      :410           
 4      :394           
 (Other):952           

Code for import in import_data/malnutrition_zambia.R.

Example 1.3: Patent Opposition

patent_url <- "https://www.uni-goettingen.de/de/document/download/66eb4eb0bc0e8f6acf1d02ddf683f077.raw/patentdata.raw"

patent_opposition <- read.table(
  url(patent_url),
  header = 1,
  colClasses = c(
    "factor", "factor", "factor",
    "factor", "factor", "integer",
    "integer", "integer", "integer"
  )
)
# Convert to logical
patent_opposition$biopharm <- patent_opposition$biopharm == 1
patent_opposition$ustwin <- patent_opposition$ustwin == 1
patent_opposition$patus <- patent_opposition$patus == 1
patent_opposition$patgsgr <- patent_opposition$patgsgr == 1
summary(patent_opposition)

 opp       biopharm         ustwin          patus          patgsgr       
 0:2847   Mode :logical   Mode :logical   Mode :logical   Mode :logical  
 1:2019   FALSE:2710      FALSE:1905      FALSE:3224      FALSE:3723     
          TRUE :2156      TRUE :2961      TRUE :1642      TRUE :1143     
                                                                         
                                                                         
                                                                         
      year           ncit          ncountry         nclaims      
 Min.   :1980   Min.   : 0.00   Min.   : 1.000   Min.   :  1.00  
 1st Qu.:1989   1st Qu.: 0.00   1st Qu.: 4.000   1st Qu.:  7.00  
 Median :1992   Median : 1.00   Median : 7.000   Median : 10.00  
 Mean   :1991   Mean   : 1.64   Mean   : 7.796   Mean   : 13.13  
 3rd Qu.:1994   3rd Qu.: 2.00   3rd Qu.:11.000   3rd Qu.: 16.00  
 Max.   :1997   Max.   :40.00   Max.   :17.000   Max.   :355.00  

R code in import_data/patent_opposition.R.

Example 1.4: Forest Health Status

forest_url <- "https://www.uni-goettingen.de/de/document/download/f5ef58e05aff8b6546dcf993aa73a480.raw/beach.raw"

forest_health_status <- read.table(
  url(forest_url),
  header = 1,
  colClasses = c(
    "factor", "numeric", "factor",
    "numeric", "numeric", "numeric",
    "numeric", "numeric", "numeric",
    "numeric", "numeric", "factor",
    "factor", "factor", "factor",
    "factor"
  ),
  na.strings = "."
)
summary(forest_health_status)

       id            year       defol            x               y        
 1      :  22   Min.   :1983   0   :1116   Min.   : 0.70   Min.   :0.400  
 10     :  22   1st Qu.:1988   12.5: 435   1st Qu.: 3.70   1st Qu.:2.100  
 11     :  22   Median :1994   25  : 126   Median : 8.10   Median :3.000  
 12     :  22   Mean   :1994   37.5:  68   Mean   : 7.46   Mean   :3.306  
 13     :  22   3rd Qu.:1999   50  :  29   3rd Qu.:10.50   3rd Qu.:4.100  
 14     :  22   Max.   :2004   62.5:  16   Max.   :16.10   Max.   :9.000  
 (Other):1664                  75  :   6                                  
      age           canopyd         gradient          alt          depth      
 Min.   :  7.0   Min.   :  0.0   Min.   : 0.00   Min.   :250   Min.   : 9.00  
 1st Qu.: 65.0   1st Qu.: 70.0   1st Qu.: 6.00   1st Qu.:340   1st Qu.:16.00  
 Median :112.0   Median : 90.0   Median :14.00   Median :390   Median :23.00  
 Mean   :106.1   Mean   : 77.3   Mean   :15.45   Mean   :387   Mean   :24.64  
 3rd Qu.:148.0   3rd Qu.:100.0   3rd Qu.:21.00   3rd Qu.:440   3rd Qu.:31.00  
 Max.   :234.0   Max.   :100.0   Max.   :46.00   Max.   :480   Max.   :51.00  
                                                                              
       ph        watermoisture alkali      humus     type    fert    
 Min.   :3.280   1:198         1:352   1      :512   0:902   0:1453  
 1st Qu.:4.100   2:990         2:991   0      :463   1:894   1: 343  
 Median :4.250   3:608         3:308   2      :388                   
 Mean   :4.295                 4:145   3      :266                   
 3rd Qu.:4.440                         4      :118                   
 Max.   :6.050                         5      : 37                   
 NA's   :3                             (Other): 12                   

R code in import_data/forest_health_status.R.

Example 1.5: Munich Rent Index - Univariate Distributions

par(mfrow = c(2, 2))

ylab <- 'estimated density'
hist(munich_rent_index$rent, freq=FALSE, xlab = 'net rent in Euro', ylab = ylab)
lines(density(munich_rent_index$rent), col = "red", lwd = 2)

hist(munich_rent_index$rentsqm, freq=FALSE, xlab = 'net rent per sqm in Euro', ylab = ylab)
lines(density(munich_rent_index$rentsqm), col = "red", lwd = 2)

hist(munich_rent_index$area, freq=FALSE, xlab = 'area ub sqm', ylab = ylab)
lines(density(munich_rent_index$area), col = "red", lwd = 2)

hist(munich_rent_index$yearc, freq=FALSE, xlab = 'year of construction', ylab = ylab)
lines(density(munich_rent_index$yearc), col = "red", lwd = 2)

Example 1.6: Malnutrition in Zambia - Univariate Distributions

par(mfrow = c(3, 2))

ylab <- 'estimated density'
hist(malnutrition_zambia$zscore, freq=FALSE, xlab = 'child\'s Z-score', ylab = ylab)
lines(density(malnutrition_zambia$zscore), col = "red", lwd = 2)

hist(malnutrition_zambia$c_breastf, freq=FALSE, xlab = 'duration of breast feeding in month', ylab = ylab)
lines(density(malnutrition_zambia$c_breastf), col = "red", lwd = 2)

hist(malnutrition_zambia$c_age, freq=FALSE, xlab = 'child\'s age in months', ylab = ylab)
lines(density(malnutrition_zambia$c_age), col = "red", lwd = 2)

hist(malnutrition_zambia$m_agebirth, freq=FALSE, xlab = 'mother\'s age at birth', ylab = ylab)
lines(density(malnutrition_zambia$m_agebirth), col = "red", lwd = 2)

hist(malnutrition_zambia$m_height, freq=FALSE, xlab = 'mother\'s height in cm', ylab = ylab)
lines(density(malnutrition_zambia$m_height), col = "red", lwd = 2)

hist(malnutrition_zambia$m_bmi, freq=FALSE, xlab = 'mother\'s BMI', ylab = ylab)
lines(density(malnutrition_zambia$m_bmi), col = "red", lwd = 2)

Example 1.7: Munich Rent Index - Scatter Plots

par(mfrow = c(2, 2))

plot(
  x = munich_rent_index$area,
  y = munich_rent_index$rent,
  xlab = "area in sqm",
  ylab = "net rent in Euro"
)

plot(
  x = munich_rent_index$area,
  y = munich_rent_index$rentsqm,
  xlab = "area in sqm",
  ylab = "net rent per sqm in Euro"
)

plot(
  x = munich_rent_index$yearc,
  y = munich_rent_index$rent,
  xlab = "year of construction",
  ylab = "net rent in Euro"
)

plot(
  x = munich_rent_index$yearc,
  y = munich_rent_index$rentsqm,
  xlab = "year of construction",
  ylab = "net rent per sqm in Euro"
)

Example 1.8: Munich Rent Index - Clusttered Scatter Plots

par(mfrow = c(2, 2))

boxplot(
  rent ~ area,
  data = munich_rent_index,
  xlab = "area in sqm",
  ylab = "net rent in Euro",
  xaxt = "n"
)

boxplot(
  rentsqm ~ area,
  data = munich_rent_index,
  xlab = "area in sqm",
  ylab = "net rent per sqm in Euro",
  xaxt = "n"
)

boxplot(
  rent ~ yearc,
  data = munich_rent_index,
  xlab = "year of construction",
  ylab = "net rent in Euro",
  xaxt = "n"
)

boxplot(
  rentsqm ~ yearc,
  data = munich_rent_index,
  xlab = "year of construction",
  ylab = "net rent per sqm in Euro",
  xaxt = "n"
)

Example 1.9: Munich Rent Index - Categorical Explanatory Variables

par(mfrow = c(1, 2))

boxplot(
  rentsqm ~ location,
  data = munich_rent_index,
  xlab = '',
  ylab = 'net rennt per sqm',
  names = c('average', 'good', 'top')
)

plot(density(subset(munich_rent_index, location == 1)$rentsqm), xlab = 'net rent per sqm', ylab = 'estimated density', lwd = 2)
lines(density(subset(munich_rent_index, location == 2)$rentsqm), col = "green", lwd= 2)
lines(density(subset(munich_rent_index, location == 3)$rentsqm), col = "red", lwd= 2)

legend(
  "topright",
  legend = c('average', 'good', 'top'),
  col = c('black', 'green', 'red'),
  lwd = 2
)

Example 1.10: Malnutrition in Zambia - Graphical Association Analysis

par(mfrow = c(1, 2))

plot(
  x = malnutrition_zambia$c_age,
  y = malnutrition_zambia$zscore,
  xlab = "child's age in month",
  ylab = "Z-score"
)

boxplot(
  zscore ~ c_age,
  data = malnutrition_zambia,
  xlab = "child's age in month",
  ylab = "Z-score",
  xaxt = "n"
)

par(mfrow = c(3, 2))

boxplot(
  zscore ~ c_breastf,
  data = malnutrition_zambia,
  xlab = "duration of breastfeeding in months",
  ylab = "average Z-score",
  xaxt = "n"
)

malnutrition_zambia$bmi_group <- cut(
  malnutrition_zambia$m_bmi,
  breaks = seq(
    min(malnutrition_zambia$m_bmi),
    max(malnutrition_zambia$m_bmi),
    length.out = 101
  ),
  include.lowest = TRUE
)

boxplot(
  zscore ~ bmi_group,
  data = malnutrition_zambia,
  xlab = "mother's BMI",
  ylab = "average Z-score",
  xaxt = "n"
)

malnutrition_zambia$age_group <- cut(
  malnutrition_zambia$m_agebirth,
  breaks = seq(
    min(malnutrition_zambia$m_agebirth),
    max(malnutrition_zambia$m_agebirth),
    length.out = 101
  ),
  include.lowest = TRUE
)

boxplot(
  zscore ~ age_group,
  data = malnutrition_zambia,
  xlab = "mother's age in years",
  ylab = "average Z-score",
  xaxt = "n"
)

malnutrition_zambia$height_group <- cut(
  malnutrition_zambia$m_height,
  breaks = seq(
    min(malnutrition_zambia$m_height),
    max(malnutrition_zambia$m_height),
    length.out = 101
  ),
  include.lowest = TRUE
)

boxplot(
  zscore ~ height_group,
  data = malnutrition_zambia,
  xlab = "mother's height in cm",
  ylab = "average Z-score",
  xaxt = "n"
)

boxplot(
  zscore ~ m_education,
  data = malnutrition_zambia,
  xlab = 'mother\'s level of education',
  ylab = 'Z-score',
  names = c('none', 'primary', 'secondary', 'higher')
)