问题1：单个向量的 density 分布图？

(1) 模拟数据

set.seed(202402)
dat=diamonds[sample(nrow(diamonds), 1000),]> head(dat)
# A tibble: 6 × 10carat cut       color clarity depth table price     x     y     z<dbl> <ord>     <ord> <ord>   <dbl> <dbl> <int> <dbl> <dbl> <dbl>
1  0.35 Ideal     I     VS2      59.8    57   630  4.6   4.59  2.75
2  0.59 Ideal     D     SI1      61.8    56  1816  5.37  5.4   3.33
3  0.39 Ideal     D     VS1      62      57  1095  4.67  4.71  2.91
4  1.12 Premium   G     IF       60.9    57  9126  6.79  6.68  4.1 
5  0.51 Very Good E     VVS2     62.1    55  2056  5.14  5.16  3.2 
6  1.03 Premium   G     VVS2     60.4    59  7729  6.58  6.56  3.97

(2) 绘制price的分布曲线

效果图：

ggplot(data = dat, mapping = aes(x=price)) + geom_histogram(aes(y=..density..),binwidth = 400,#bins=30,fill="bisque",color="white",alpha=0.7) + geom_density() +geom_rug() +labs(x='Price') +theme_minimal(base_size = 12)

问题2：按照A的值分bin统计B的平均值，并画图

求不同克拉区间的钻石，其价格的分布及平均值。

• 对变量 carat 分区间
• 统计每个 carat 区间的 price 价格的平均值
• 画图

(1) 使用cut函数对carat列划分到10个区间中，求每个区间price均值

效果图：

重要函数：cut(x, breaks=n) 对连续向量x分为n个bin，返回每个x值对应的bin范围，因子型。

> # 按照A的值分bin统计B的平均值，并画图 ----
> dat$tags = cut(dat$carat, breaks = 10) #分10个区间
> head(dat)
# A tibble: 6 × 11carat cut       color clarity depth table price     x     y     z tags         <dbl> <ord>     <ord> <ord>   <dbl> <dbl> <int> <dbl> <dbl> <dbl> <fct>        
1  0.35 Ideal     I     VS2      59.8    57   630  4.6   4.59  2.75 (0.217,0.499]
2  0.59 Ideal     D     SI1      61.8    56  1816  5.37  5.4   3.33 (0.499,0.778]
3  0.39 Ideal     D     VS1      62      57  1095  4.67  4.71  2.91 (0.217,0.499]
4  1.12 Premium   G     IF       60.9    57  9126  6.79  6.68  4.1  (1.06,1.34]  
5  0.51 Very Good E     VVS2     62.1    55  2056  5.14  5.16  3.2  (0.499,0.778]
6  1.03 Premium   G     VVS2     60.4    59  7729  6.58  6.56  3.97 (0.778,1.06] 求每个carat区间中price的平均值
> results=sapply(split(dat$price, dat$tags), function(x){
+   mean(x)
+ })
> levels(dat$tags)[1] "(0.217,0.499]" "(0.499,0.778]" "(0.778,1.06]"  "(1.06,1.34]"   "(1.34,1.61]"   "(1.61,1.89]"   "(1.89,2.17]"  [8] "(2.17,2.45]"   "(2.45,2.73]"   "(2.73,3.01]"  
> head(results)
(0.217,0.499] (0.499,0.778]  (0.778,1.06]   (1.06,1.34]   (1.34,1.61]   (1.61,1.89] 793.9373     2105.2756     4961.7216     6500.9048     9756.3146    10490.7308 
> #results[levels(dat$tags)]
> plot( as.numeric( results), type="o", pch=19,
+       xlab="bin index", ylab="Price", mgp=c(2,1,0))
> 
> 

(2) 绘制每个carat bin的price的小提琴图

效果图：

代码：

数据同上(1)：> summary(dat)carat               cut      color      clarity        depth           table           price             x        Min.   :0.2200   Fair     : 35   D:129   SI1    :227   Min.   :55.90   Min.   :52.00   Min.   :  345   Min.   :3.900  1st Qu.:0.4000   Good     : 89   E:178   VS2    :220   1st Qu.:61.10   1st Qu.:56.00   1st Qu.:  987   1st Qu.:4.720  Median :0.7000   Very Good:206   F:181   SI2    :173   Median :61.90   Median :57.00   Median : 2362   Median :5.680  Mean   :0.8033   Premium  :254   G:203   VS1    :156   Mean   :61.82   Mean   :57.42   Mean   : 3939   Mean   :5.745  3rd Qu.:1.0600   Ideal    :416   H:136   VVS2   :104   3rd Qu.:62.60   3rd Qu.:59.00   3rd Qu.: 5443   3rd Qu.:6.560  Max.   :3.0100                   I:119   VVS1   : 62   Max.   :79.00   Max.   :73.00   Max.   :18432   Max.   :9.540  J: 54   (Other): 58                                                                  y               z                    tags    Min.   :3.880   Min.   :2.310   (0.217,0.499]:319  1st Qu.:4.740   1st Qu.:2.920   (0.499,0.778]:254  Median :5.700   Median :3.520   (0.778,1.06] :176  Mean   :5.747   Mean   :3.552   (1.06,1.34]  :105  3rd Qu.:6.560   3rd Qu.:4.050   (1.34,1.61]  : 89  Max.   :9.380   Max.   :5.900   (1.61,1.89]  : 26  (Other)      : 31  

绘图代码：

library(ggplot2)
ggplot(dat, aes(x=tags, y=price, fill=tags))+geom_violin(scale="width", color="#00112200")+geom_boxplot(width=0.1, fill="white", outlier.size = 0.1)+geom_jitter(color="blue", alpha=0.2, size=0.5, shape=19)+theme_classic(base_size = 14)+#coord_flip()+theme(axis.text.x=element_text(angle=30, hjust=1),legend.position = "none",)+labs(x="carat", y="price")+#ylim(0.5, 0.85)+scale_fill_manual(values= c(scales::hue_pal()(12)) )