R：如何自动计算数据框，然后根据多个数据框的结果生成图表？

Question

How can I build a dynamic "downflow pipeline" to push data frames through with R to auto-calculate formulas using these data frames? I have this data frame called autocalc, which has blank spaces carved out for equations. For example, I need to apply equations such as this: autocalc$PPH <- Tokyo$P / Tokyo$PH . PPH is already a column/vector.

autocalc <- data.frame("INCOME" = c("$0 to $15,000","$15,000 to $29,999","$30,000 to $39,999","$40,000 to $49,999","$50,000 to $69,999","$70,000 to $99,999","$100,000 to $149,999"),
                             "PPH" = c(0,0,0,0,0,0,0),
                             "PTS" = c(0,0,0,0,0,0,0))
    
    autocalc$PPH <- Tokyo$P / Tokyo$PH 
    autocalc$PTS <- autocalc$PPH * .05

The results of this data frame will be used to build a chart with ggplot2.

As you can see from the sample equation, there is a data frame called Tokyo. I have three city data frames (NewYork, Paris, Tokyo). Each data frame has the city's prisoners (vector P) and population (vector PH). The data is further categorized by income group.

Tokyo <- data.frame("INCOME" = c("$0 to $15,000","$15,000 to $29,999","$30,000 to $39,999","$40,000 to $49,999","$50,000 to $69,999","$70,000 to $99,999","$100,000 to $149,999"),
                          "P" = c(1844,1062,1036,448,770,364,395),
                          "PH" = c(84900,721007,80800,380004,675000,32900,39500))

I want to apply each city to the autocalc data frame and produce three separate charts. I could copy/paste three versions of autocalc for each city, but that seems like bad code because if I have to fix something in autocalc, I have to redo the code for each city.

I looked at the following leads, but I feel like this shouldn't be that complex of an issue:

1. Use objective oriented programming with the R6 library. Create an autocalc class and apply a copy to the three cities? There aren't many R6 tutorials that make sense for a beginner, so I feel like there's an easier way?

2. Use lapply(). I found a tutorial that produces something similar (three separate charts based on three separate data frames), but it can't explain how to swap autocalc$PPH <- Tokyo$P / Tokyo$PH for autocalc$PPH <- Paris$P / Paris$PH . And to preserve autocalc's calculations with each city, to prepare three charts. Should I instead get rid of the autocalc data frame and add rows to each city's data frame?

3. My friend not well versed in R, but is a programmer, and recommends looking into collection methods. To write a script and use methods to do repeated operations. However, I can't find a tutorial to do with R. I think this requires me to use object-oriented programming with the R6 library? I'm thinking maybe my task is better suited with Python then? I'm being told that R is more for analysis and not building something as dynamic as this.

Answer 1

I think option 2 is the most easy and straight-forward one. You can put the 3 dataframes in a list and use lapply . You can pass an anonymous function in lapply to refer to the each cities dataframe inside the function.

list_dfs <- list(Tokyo, Paris, NewYork)

list_plots <- lapply(list_dfs, function(x) {
  autocalc$PPH <- x$P / x$PH 
  autocalc$PTS <- autocalc$PPH * .05
  #ggplot2  code here
})

After you enter the ggplot2 code in lapply , it will generate a list of 3 plots for each city which you can access by doing list_plots[[1]] , list_plots[[2]] and list_plots[[3]] .