1_4_analysedonnee

Formation R

Analyse descriptive de données

Workflow d’un projet Data Science

EDA

Comprendre son dataset et ses caractéristiques

décrire les variables
visualiser les variables
analyser leurs relations

Calcul de statistiques descriptives

summary()

library(readr)

athome <- read_csv( 'data/athome.csv') 

summary(athome)

##        id              img                link              source         
##  Min.   : 199285   Length:144677      Length:144677      Length:144677     
##  1st Qu.:6532352   Class :character   Class :character   Class :character  
##  Median :6682461   Mode  :character   Mode  :character   Mode  :character  
##  Mean   :6621094                                                           
##  3rd Qu.:6733616                                                           
##  Max.   :6788549                                                           
##                                                                            
##     title              terrain        type_annonce        type_bien        
##  Length:144677      Min.   :   0.01   Length:144677      Length:144677     
##  Class :character   1st Qu.:   1.81   Class :character   Class :character  
##  Mode  :character   Median :   2.02   Mode  :character   Mode  :character  
##                     Mean   :   2.52                                        
##                     3rd Qu.:   2.02                                        
##                     Max.   :4000.00                                        
##                     NA's   :75944                                          
##  type_bien_dtl          pieces       annee_de_construction   quartier        
##  Length:144677      Min.   : 1.00    Min.   :       6      Length:144677     
##  Class :character   1st Qu.: 3.00    1st Qu.:    2010      Class :character  
##  Mode  :character   Median : 5.00    Median :    2014      Mode  :character  
##                     Mean   : 5.25    Mean   :    2175                        
##                     3rd Qu.: 7.00    3rd Qu.:    2019                        
##                     Max.   :25.00    Max.   :20202021                        
##                     NA's   :143713   NA's   :8949                            
##     ville               etage             ascenseur      salles_de_bains 
##  Length:144677      Min.   :1.000e+00   Min.   :0.0000   Min.   : 1.00   
##  Class :character   1st Qu.:1.000e+00   1st Qu.:0.0000   1st Qu.: 1.00   
##  Mode  :character   Median :2.000e+00   Median :0.0000   Median : 1.00   
##                     Mean   :3.711e+05   Mean   :0.4866   Mean   : 1.08   
##                     3rd Qu.:3.000e+00   3rd Qu.:1.0000   3rd Qu.: 1.00   
##                     Max.   :2.008e+09   Max.   :1.0000   Max.   :14.00   
##                     NA's   :139266                       NA's   :104315  
##  salles_de_douches   wc_separes        terrasse          balcon       
##  Min.   : 1.00     Min.   :0.0000   Min.   :0.0000   Min.   :0.00000  
##  1st Qu.: 1.00     1st Qu.:0.0000   1st Qu.:1.0000   1st Qu.:0.00000  
##  Median : 1.00     Median :0.0000   Median :1.0000   Median :0.00000  
##  Mean   : 1.01     Mean   :0.4946   Mean   :0.9457   Mean   :0.01895  
##  3rd Qu.: 1.00     3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:0.00000  
##  Max.   :30.00     Max.   :1.0000   Max.   :1.0000   Max.   :1.00000  
##  NA's   :107583                                                       
##      jardin            cave          buanderie         grenier 
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :1  
##  1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:1  
##  Median :0.0000   Median :0.0000   Median :0.0000   Median :1  
##  Mean   :0.4779   Mean   :0.4961   Mean   :0.4935   Mean   :1  
##  3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:1.0000   3rd Qu.:1  
##  Max.   :1.0000   Max.   :1.0000   Max.   :1.0000   Max.   :1  
##                                                                
##     cheminee        cuisine_ouverte  cuisine_equipee   panneaux_solaires
##  Min.   :0.000000   Min.   :0.0000   Min.   :0.00000   Min.   :0.0000   
##  1st Qu.:0.000000   1st Qu.:0.0000   1st Qu.:0.00000   1st Qu.:0.0000   
##  Median :0.000000   Median :0.0000   Median :0.00000   Median :0.0000   
##  Mean   :0.003553   Mean   :0.2436   Mean   :0.02339   Mean   :0.2358   
##  3rd Qu.:0.000000   3rd Qu.:0.0000   3rd Qu.:0.00000   3rd Qu.:0.0000   
##  Max.   :1.000000   Max.   :1.0000   Max.   :1.00000   Max.   :1.0000   
##                                                                         
##  acces_mobilite_reduite charges_mensuelles    caution           meuble      
##  Min.   :0.000000       Min.   :     1.0   Min.   :     1   Min.   :0.0000  
##  1st Qu.:0.000000       1st Qu.:   200.0   1st Qu.:  2900   1st Qu.:0.0000  
##  Median :0.000000       Median :   200.0   Median :  2900   Median :0.0000  
##  Mean   :0.003325       Mean   :   208.3   Mean   :  2964   Mean   :0.2436  
##  3rd Qu.:0.000000       3rd Qu.:   200.0   3rd Qu.:  2900   3rd Qu.:0.0000  
##  Max.   :1.000000       Max.   :100000.0   Max.   :150000   Max.   :1.0000  
##                         NA's   :75730      NA's   :76808                    
##  chauffage_gaz    chauffage_electrique chauffage_geothermique
##  Min.   :0.0000   Min.   :0.000000     Min.   :0.0000000     
##  1st Qu.:0.0000   1st Qu.:0.000000     1st Qu.:0.0000000     
##  Median :0.0000   Median :0.000000     Median :0.0000000     
##  Mean   :0.0344   Mean   :0.000235     Mean   :0.0005046     
##  3rd Qu.:0.0000   3rd Qu.:0.000000     3rd Qu.:0.0000000     
##  Max.   :1.0000   Max.   :1.000000     Max.   :1.0000000     
##                                                              
##  chauffage_mazout   montant_agence      piscine          place_parking   
##  Min.   :0.000000   Min.   :    0    Min.   :0.0000000   Min.   :0.0000  
##  1st Qu.:0.000000   1st Qu.: 1170    1st Qu.:0.0000000   1st Qu.:1.0000  
##  Median :0.000000   Median : 1755    Median :0.0000000   Median :1.0000  
##  Mean   :0.002737   Mean   : 2206    Mean   :0.0005322   Mean   :0.9573  
##  3rd Qu.:0.000000   3rd Qu.: 2500    3rd Qu.:0.0000000   3rd Qu.:1.0000  
##  Max.   :1.000000   Max.   :20475    Max.   :1.0000000   Max.   :1.0000  
##                     NA's   :143614                                       
##  parking_interieur  parking_exterieur     chambres         surface        
##  Min.   :0.00e+00   Min.   :0.000000   Min.   : 0.000   Min.   :     9.0  
##  1st Qu.:0.00e+00   1st Qu.:0.000000   1st Qu.: 2.000   1st Qu.:    98.0  
##  Median :0.00e+00   Median :0.000000   Median : 2.000   Median :   105.0  
##  Mean   :8.99e-05   Mean   :0.002426   Mean   : 2.504   Mean   :   147.3  
##  3rd Qu.:0.00e+00   3rd Qu.:0.000000   3rd Qu.: 3.000   3rd Qu.:   190.0  
##  Max.   :1.00e+00   Max.   :1.000000   Max.   :18.000   Max.   :566818.0  
##                                        NA's   :4157     NA's   :3810      
##       prix         
##  Min.   :       1  
##  1st Qu.:  700000  
##  Median :  700000  
##  Mean   :  674517  
##  3rd Qu.:  700000  
##  Max.   :16700000  
##  NA's   :103049

summary() ne permet pas la création d’une table

Dataset de statistiques descriptives détaillées

utiliser summarise() (manuel)
utiliser skim() (automatique)

Dataset de statistiques descriptives détaillées

utiliser summarise() (manuel)
utiliser skim() (automatique)

summarise()

library(dplyr)

 athome %>% 
  filter(ville=='Luxembourg') %>% 
  group_by(type_annonce) %>% 
  summarise(max_prix=max(prix, na.rm = TRUE), min_prix=min(prix, na.rm = TRUE), mean_prix=mean(prix, na.rm = TRUE), max_surface=max(surface, na.rm = TRUE), min_surface=min(surface, na.rm = TRUE), mean_surface=mean(surface, na.rm = TRUE))

type_annonce	max_prix	min_prix	mean_prix	max_surface	min_surface	mean_surface
location	125048	1	3057.511	10000	9	90.37613
vente	16700000	1	1241678.888	800	12	103.53956

skim()

library(skimr)
library(dplyr)

 athome %>% 
  filter(ville=='Luxembourg') %>% 
  select(type_annonce, prix, charges_mensuelles, annee_de_construction, pieces, surface) %>%  
  group_by(type_annonce) %>% 
  skim() #%>%
  # yank("numeric") possibilité de sortir uniquement les numeriques ou charactere

Variable type: numeric

skim_variable	type_annonce	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
prix	location	969	0.67	3057.51	6047.28	1	850.00	1650.0	2787.00	125048	▇▁▁▁▁
prix	vente	653	0.63	1241678.89	1019439.87	1	738116.00	1050000.0	1386428.75	16700000	▇▁▁▁▁
charges_mensuelles	location	1380	0.53	258.46	283.55	1	150.00	200.0	275.00	4253	▇▁▁▁▁
charges_mensuelles	vente	1667	0.06	348.12	531.37	30	197.50	250.0	315.00	4750	▇▁▁▁▁
annee_de_construction	location	2157	0.26	1999.80	27.22	1750	1988.00	2013.0	2018.00	2020	▁▁▁▁▇
annee_de_construction	vente	1287	0.28	1993.74	129.12	12	1988.75	2013.5	2020.00	2107	▁▁▁▁▇
pieces	location	2694	0.07	4.10	2.85	1	2.00	3.0	5.00	16	▇▂▁▁▁
pieces	vente	1644	0.08	4.78	2.87	1	3.00	4.0	6.00	15	▇▇▃▁▁
surface	location	1326	0.54	90.38	261.42	9	45.00	70.0	97.00	10000	▇▁▁▁▁
surface	vente	149	0.92	103.54	63.78	12	62.53	86.0	122.48	800	▇▁▁▁▁

Tableau de contingence

 select_athome <- athome %>% 
  filter(ville=='Luxembourg' & quartier %in% c('Beggen', 'Eich', 'Belair', 'Merl', 'Kirchberg')) %>% 
  select(type_annonce, quartier) 

table(select_athome) %>% 
    as.data.frame.matrix

Visualiser les prix des locations par quartier

Package `forcats`

Avec forcats, il est possible de manipuler les variables ordinales

 select_athome2 <- select_athome %>% 
  mutate(quartier = factor(quartier, levels=c('Belair', 'Merl', 'Kirchberg','Beggen', 'Eich') ))
           
# inverser les "levels" 
 select_athome3 <- select_athome2 %>% 
  mutate(quartier = fct_rev(quartier) )

# associer l'ordre à une autre variable
 select_athome4 <-  select_athome %>% 
  mutate(ord = case_when(quartier =='Belair' ~ 1, 
                         quartier =='Merl' ~ 5, 
                         quartier =='Kirchberg' ~ 2,
                         quartier =='Beggen' ~4, 
                         quartier =='Eich' ~ 3),
         quartier = fct_reorder(quartier, ord))

pop <- pop %>% 
         mutate(county=fct_rev(county))

Code associé

library(ggridges)
library(ggplot2)
library(hrbrthemes)
library(viridis)
library(scales)

 select_athome <- athome %>% 
  filter(type_annonce=='location' & quartier %in% c('Beggen', 'Eich', 'Belair', 'Merl', 'Kirchberg')) %>% 
  select(prix, quartier) 

ggplot(select_athome, aes(x = prix, y = quartier, fill = quartier))+
  geom_density_ridges(alpha=0.6, stat="binline", bins=20) +
  scale_fill_viridis(discrete=TRUE) +
  scale_color_viridis(discrete=TRUE) +
  theme_ipsum() +
  theme(
    legend.position="none",
    panel.spacing = unit(0.1, "lines"),
    strip.text.x = element_text(size = 8)
  ) + 
  scale_x_continuous(labels = comma) +
  xlab("Prix des locations") +
  ylab(" ")

1.4.1 Instructions

1) Allez dans acc2019
2) Tableau descriptif des variables intéressantes
3) Tableau de contengence avec espece et domcor
4) Visualiser sous forme d’histogramme les accidents par rapport à l’heure et à la vitesse associée à la zone

Pondération dans votre analyse

{Hmisc} facilite la pondération :

library(Hmisc)

x  # la variable d'intérêt
weights # le poids associé

wtd.quantile(x, weights)  # weighted quantile
wtd.mean(x, weights)      # weighted mean
wtd.var(x, weights)       # weighted variance

Contact

Kevin Rosamont Prombo

Formation R

Workflow d’un projet Data Science

EDA

Calcul de statistiques descriptives

summary()

Dataset de statistiques descriptives détaillées

Dataset de statistiques descriptives détaillées

summarise()

skim()

Tableau de contingence

Visualiser les prix des locations par quartier

Package forcats

Code associé

1.4.1 Instructions

Pondération dans votre analyse

Contact

Package `forcats`