调整Secondary Axis匹配Primary Axis：实现次轴起始于0的方法

需求说明 #

• 主轴（Avg. Asking Rate）始终使用自身数据的最小值作为下限
• 次轴（Vacancy Rate）强制从0开始，不允许出现负值

问题根源 #

原代码的train_sec函数使用了次轴数据的完整范围（range(secondary)）进行缩放转换，导致次轴下限会跟随原始数据的最小值变动，无法固定为0。

修改方案 #

调整次轴的缩放基准范围，强制将次轴的转换起始点设为0，而非原始数据的最小值。以下是修改后的完整代码：

library(dplyr)   
library(scales)  
library(ggplot2)

# 示例数据
stats <- data.frame(
  Snapshot.Date = as.Date(c("2020-09-30", "2020-12-31", "2021-03-31", "2021-06-30", "2021-09-30", "2021-12-31")),
  avg_asking_rate = c(
    4.93799293627304, 4.99132174488857, 4.97053998470495, 4.90321494004932,
    4.93494348159945, 4.91379691657121
  ),
  vacancy_rate = c(
    9.94185919399802, 11.5989188611645, 12.9308555110482, 13.9590437469142,
    13.5155036299108, 12.8472483323431
  )
)

# 修改后的次轴转换函数：强制次轴从0开始
train_sec <- function(primary, secondary, na.rm = TRUE) {
  # 次轴范围改为从0到最大值
  from <- c(0, max(secondary, na.rm = na.rm))
  to   <- range(primary, na.rm = na.rm)
  forward <- function(x) {
    rescale(x, from = from, to = to)
  }
  reverse <- function(x) {
    rescale(x, from = to, to = from)
  }
  list(fwd = forward, rev = reverse)
}

# 主轴下限取0.5的倍数（保持原逻辑）
y_lm <- floor((min(stats$avg_asking_rate))*2)/2

# 生成转换规则
sec <- with(stats, train_sec(stats$avg_asking_rate, stats$vacancy_rate))

# 绘图
stats %>%
  ggplot(., aes(x = `Snapshot.Date`, y = avg_asking_rate))+
  ggtitle("Rent") +
  geom_col(aes(fill = "Avg. Asking Rate"))+
  geom_line(aes(y = sec$fwd(vacancy_rate), colour =  "Vacancy Rate"), size = 1) +
  
  scale_x_date(name = NULL)+
  scale_y_continuous(name = "",
                     sec.axis = sec_axis(~sec$rev(.), name = "Vacancy Rate"))+
  coord_cartesian(ylim=c(y_lm,NA)) +
  scale_fill_manual(name = NULL, values = c("Avg. Asking Rate" = "blue", "Vacancy Rate" = "green"),
                    guide = guide_legend(order = 1))+
  scale_colour_manual(name = NULL, values = c("Vacancy Rate" = "green"), guide = guide_legend())

关键修改点 #

• 在train_sec函数中，将次轴的缩放范围from从range(secondary)改为c(0, max(secondary, na.rm=na.rm))，强制次轴以0为下限
• 主轴的下限逻辑保持不变，仍然基于自身数据最小值取0.5的倍数
• 次轴的反向转换会自动基于新的范围生成，确保刻度从0开始

预期效果 #

内容的提问来源于stack exchange，提问作者wigglesthe3rd