Data wrangling
As you can see not all formats of our variables are adjusted. We need to prepare the appropriate formats of our variables according to their measurement scales and future usage.
mieszkania$district<-as.factor(mieszkania$district)
mieszkania$building_type<-as.factor(mieszkania$building_type)
mieszkania$rooms<-factor(mieszkania$rooms,ordered=TRUE)
attach(mieszkania)
mieszkania$price_PLN<-as.numeric(mieszkania$price_PLN)
mieszkania$price_EUR<-as.numeric(mieszkania$price_EUR)Frequency table
#PLN
Range1<-cut(price_PLN,seq(350000,1350000,by=100000))
table1<-table(Range1)
transform(table1,Prop_Freq=prop.table(Freq),Cum_Freq=cumsum(Freq))## Range1 Freq Prop_Freq Cum_Freq
## 1 (3.5e+05,4.5e+05] 9 0.045 9
## 2 (4.5e+05,5.5e+05] 21 0.105 30
## 3 (5.5e+05,6.5e+05] 33 0.165 63
## 4 (6.5e+05,7.5e+05] 36 0.180 99
## 5 (7.5e+05,8.5e+05] 31 0.155 130
## 6 (8.5e+05,9.5e+05] 36 0.180 166
## 7 (9.5e+05,1.05e+06] 21 0.105 187
## 8 (1.05e+06,1.15e+06] 10 0.050 197
## 9 (1.15e+06,1.25e+06] 2 0.010 199
## 10 (1.25e+06,1.35e+06] 1 0.005 200#EUR
Range2<-cut(price_EUR,seq(80000,320000,by=40000))
table2<-table(Range2)
transform(table2,Prop_Freq=prop.table(Freq),Cum_Freq=cumsum(Freq))## Range2 Freq Prop_Freq Cum_Freq
## 1 (8e+04,1.2e+05] 20 0.100 20
## 2 (1.2e+05,1.6e+05] 59 0.295 79
## 3 (1.6e+05,2e+05] 54 0.270 133
## 4 (2e+05,2.4e+05] 52 0.260 185
## 5 (2.4e+05,2.8e+05] 13 0.065 198
## 6 (2.8e+05,3.2e+05] 2 0.010 200TAI
## # classes Goodness of fit Tabular accuracy
## 10.0000000 0.9780872 0.8508467## # classes Goodness of fit Tabular accuracy
## 6.0000000 0.9348720 0.7429525Basic plots
In this section we should present our data using basic (pre-installed with R) graphics. Choose the most appropriate plots according to the scale of chosen variables. Investigate the heterogeneity of the distribution presenting data by groups (i.e. by district, building type etc.). Do not forget about main titles, labels and legend. Read more about graphical parameters here.
Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.
ggplot2 plots
Using facets
Faceting generates small multiples each showing a different subset of the data. Small multiples are a powerful tool for exploratory data analysis: you can rapidly compare patterns in different parts of the data and see whether they are the same or different. Read more about facets here.
Descriptive statistics #1
Before automatically reporting the full summary table of descriptive statistics, this time your goal is to measure the central tendency of the distribution of prices. Compare mean, median and mode together with positional measures - quantiles - by districts and building types or no. of rooms per apartment.
## [1] "PLN"## [1] 760035## [1] 755719.5## [1] 186099.8## [1] 34633125960## 75%
## 282686.5## [1] 359769## [1] 1277691## 0% 5% 25% 50% 75% 95% 100%
## 359769.0 477175.4 619073.8 755719.5 901760.2 1054250.8 1277691.0## [1] "EUR"## [1] 175934## [1] 174935## [1] 43078.62## [1] 1855767906## 75%
## 65436.25## [1] 83280## [1] 295762## 0% 5% 25% 50% 75% 95% 100%
## 83280.0 110457.8 143304.2 174935.0 208740.5 244039.9 295762.0Summary tables with ‘kable’
Using kable and kableextra packages we can easily create summary tables with graphics and/or statistics.
#PLN
mieszkania_list <- split(mieszkania$price_PLN, mieszkania$rooms)
inline_plot <- data.frame(rooms = c(1, 2, 3, 4), boxplot = "", histogram = "",
line1 = "", line2 = "", points1 = "")
inline_plot %>%
kbl(booktabs = TRUE) %>%
kable_paper(full_width = FALSE) %>%
column_spec(2, image = spec_boxplot(mieszkania_list)) %>%
column_spec(3, image = spec_hist(mieszkania_list)) %>%
column_spec(4, image = spec_plot(mieszkania_list, same_lim = TRUE)) %>%
column_spec(5, image = spec_plot(mieszkania_list, same_lim = FALSE)) %>%
column_spec(6, image = spec_plot(mieszkania_list, type = "p"))| rooms | boxplot | histogram | line1 | line2 | points1 |
|---|---|---|---|---|---|
| 1 | |||||
| 2 | |||||
| 3 | |||||
| 4 |
#EUR
mieszkania_list <- split(mieszkania$price_EUR, mieszkania$rooms)
inline_plot <- data.frame(rooms = c(1, 2, 3, 4), boxplot = "", histogram = "",
line1 = "", line2 = "", points1 = "")
inline_plot %>%
kbl(booktabs = TRUE) %>%
kable_paper(full_width = FALSE) %>%
column_spec(2, image = spec_boxplot(mieszkania_list)) %>%
column_spec(3, image = spec_hist(mieszkania_list)) %>%
column_spec(4, image = spec_plot(mieszkania_list, same_lim = TRUE)) %>%
column_spec(5, image = spec_plot(mieszkania_list, same_lim = FALSE)) %>%
column_spec(6, image = spec_plot(mieszkania_list, type = "p"))| rooms | boxplot | histogram | line1 | line2 | points1 |
|---|---|---|---|---|---|
| 1 | |||||
| 2 | |||||
| 3 | |||||
| 4 |
Ok, now we will finally summarize basic central tendency measures for prices by districts/building types using kable packages. You can customize your final report. See some hints here.
library(psych)
#PLN and district
summary1 <-
list("price PLN" =
list(
"price PLN_min" = ~ min(price_PLN),
"price PLN_max" = ~ max(price_PLN),
"price PLN_mean(sd)" = ~ qwraps2::mean_sd(price_PLN),
"Q1" = ~ quantile(price_PLN,0.25),
"Median" = ~ median(price_PLN),
"Q3" = ~ quantile(price_PLN,0.75),
"IQR" = ~ iqr(price_PLN),
"Var %" = ~ round(sd(price_PLN)/mean(price_PLN),1),
"IQR Var %" = ~ round(iqr(price_PLN)/median(price_PLN),1),
"Skewness" = ~ round(skew(price_PLN),1),
"Kurtosis" = ~ round(kurtosis(price_PLN),1)
)
)
summary_table(mieszkania, summary1, by=c("district"))%>%
kbl(digits = 2,
caption="Table 1. prices_PLN and district",
col.names = c('Biskupin', 'Krzyki', 'Srodmiescie'))%>%
kable_classic(full_width = F, html_font = "Cambria")%>%
kable_styling(bootstrap_options = c("striped", "hover"))| Biskupin | Krzyki | Srodmiescie | |
|---|---|---|---|
| price PLN_min | 519652 | 359769 | 448196 |
| price PLN_max | 1277691 | 1090444 | 1062054 |
| price PLN_mean(sd) | 818,614.06 \(\pm\) 175,597.94 | 726,507.20 \(\pm\) 195,015.45 | 739,339.70 \(\pm\) 171,428.11 |
| Q1 | 676751 | 600180.5 | 592287.75 |
| Median | 817736 | 716726 | 727477.5 |
| Q3 | 926474 | 876306.5 | 870752.5 |
| IQR | 249723 | 276126 | 278464.75 |
| Var % | 0.2 | 0.3 | 0.2 |
| IQR Var % | 0.3 | 0.4 | 0.4 |
| Skewness | 0.3 | 0.1 | 0.1 |
| Kurtosis | -0.3 | -1 | -1.1 |
#EUR and building type
summary2 <-
list("price EUR" =
list(
"price EUR_min" = ~ min(price_EUR),
"price EUR_max" = ~ max(price_EUR),
"price EUR_mean" = ~ qwraps2::mean_sd(price_EUR),
"Q1" = ~ quantile(price_EUR,0.25),
"Median" = ~ round(median(price_EUR),2),
"Q3" = ~ quantile(price_EUR,0.75),
"IQR" = ~ iqr(price_EUR),
"Var %" = ~ round(sd(price_EUR)/mean(price_EUR),1),
"IQR Var %" = ~ round(iqr(price_EUR)/median(price_EUR),1),
"Skewness" = ~ round(skew(price_EUR),1),
"Kurtosis" = ~ round(kurtosis(price_EUR),1)
)
)
summary_table(mieszkania, summary2,by=c("building_type"))%>%
kbl(digits = 2,
caption="Table 1. prices_EUR and building type",
col.names = c('kamienica', 'niski blok', 'wiezowiec'))%>%
kable_classic(full_width = F, html_font = "Cambria")%>%
kable_styling(bootstrap_options = c("striped", "hover"))| kamienica | niski blok | wiezowiec | |
|---|---|---|---|
| price EUR_min | 96258 | 114905 | 83280 |
| price EUR_max | 284918 | 295762 | 252418 |
| price EUR_mean | 178,317.70 \(\pm\) 42,682.42 | 188,790.90 \(\pm\) 40,831.08 | 163,363.18 \(\pm\) 42,246.11 |
| Q1 | 149944 | 160399.5 | 128657.5 |
| Median | 185346 | 187013 | 157107.5 |
| Q3 | 207450 | 217558.5 | 201563 |
| IQR | 57506 | 57159 | 72905.5 |
| Var % | 0.2 | 0.2 | 0.3 |
| IQR Var % | 0.3 | 0.3 | 0.5 |
| Skewness | 0 | 0.2 | 0.2 |
| Kurtosis | -0.6 | -0.4 | -1 |
#facet: price per room of different buildings in different district
#different room size use different colour then will find the same size may have similar price per room
plot <- ggplot(mieszkania, aes(x=price_PLN, y=price_PLN/as.numeric(rooms))) +
geom_point(shape=21,colour=rooms,stroke=0.25,alpha=0.8,size=as.numeric(rooms))
plot + facet_grid(district~building_type, scales = "free", space = "free") Interpretation: From the kable report 2 we can see that there are three different districts and three different building types. From perspective of district, the minimum, maximum, mean±sd of price_PLN in Biskupin is the most expensive. All three district have skewness bigger than 0, it means the distribute will be more concentrated in the low price area. The Kurtosis are all smaller than 0. Generally, kutosis use standard of 3 which is the value of normal distribution. However, it fix it to 0 here. So it can smaller than 0. And when it smaller than 0, it means the distribution will be more average. From the perspective of building type. It is a little strange that the niski blok looks the most expensivem. But we after we take the room size into consideration, it will be a little bit different(see details in next paragraphs). The skewness of kamienica is 0 which is the standard skewness.And others are both bigger than 0. The kurtosis are also smaller than 0. Fianlly, the last graph shows that the price in Biskupln is erally expensive. But the price of niski blok will only be most expensive in Srodmiescie. In other district, it will be different. We can also find that the apartment with less room, it will be more expensive. And the price of same size of room it will be similar, because the circle with same colour are staying together.