R模型构建学习2-解决质量差的钻石反而价格贵的问题

生信编程日常

发布于 2020-04-01 16:01:19

5490

发布于 2020-04-01 16:01:19

上次的代码：

library(ggplot2)
data(diamonds)

ggplot(diamonds, aes(cut, price)) + geom_boxplot()
ggplot(diamonds, aes(color, price)) + geom_boxplot()
ggplot(diamonds, aes(clarity, price)) + geom_boxplot()

ggplot(diamonds, aes(carat, price)) + geom_hex(bins = 50)


library(tidyverse)
library(modelr)
library(magrittr)
diamonds2 <- diamonds %>%
filter(carat <= 2.5) %>%
mutate(lprice = log2(price), lcarat = log2(carat))


ggplot(diamonds2, aes(lcarat, lprice)) +
geom_hex(bins = 50)

最后得到了重量和价格的关系图：

caret vs price

拟合线性模型：

mod_diamond <- lm(lprice ~ lcarat, data = diamonds2)

然后进行反向变换，还原数据：

grid <- diamonds2 %>%
data_grid(carat = seq_range(carat, 20)) %>%
mutate(lcarat = log2(carat)) %>%
add_predictions(mod_diamond, "lprice") %>%
mutate(price = 2 ^ lprice)
ggplot(diamonds2, aes(carat, price)) +
geom_hex(bins = 50) +
geom_line(data = grid, color = "red", size = 1)

caret vs price 2

检查一下残差，用残差来代替原来的price进行预测：

diamonds2 <- diamonds2 %>%
add_residuals(mod_diamond, "lresid")
ggplot(diamonds2, aes(lcarat, lresid)) +
geom_hex(bins = 50)

residuals

ggplot(diamonds2, aes(cut, lresid)) + geom_boxplot()
ggplot(diamonds2, aes(color, lresid)) + geom_boxplot()
ggplot(diamonds2, aes(clarity, lresid)) + geom_boxplot()