Create binned variable from results of class interval determination

问题

I want to create a binned variable out of a continuous variable. I want 10 bins, with break points set from whatever results from a jenks classification. How do I assign each value to one of these 10 bins?

# dataframe w/ values (AllwdAmt)
df <- structure(list(X = c(2078L, 2079L, 2080L, 2084L, 2085L, 2086L, 
2087L, 2092L, 2093L, 2094L, 2095L, 4084L, 4085L, 4086L, 4087L, 
4088L, 4089L, 4091L, 4092L, 4093L, 4094L, 4095L, 4096L, 4097L, 
4098L, 4099L, 4727L, 4728L, 4733L, 4734L, 4739L, 4740L, 4741L, 
4742L, 4743L, 4744L, 4745L, 4746L, 4747L, 4748L, 4749L, 4750L, 
4751L, 4752L, 4753L, 4754L, 4755L, 4756L, 4757L, 4758L), AllwdAmt = c(34.66, 
105.56, 105.56, 473.93, 108, 1669.23, 201.5, 62.67, 61.54, 601.28, 
236.96, 108, 40.28, 29.32, 483.6, 236.96, 6072.4, 25.97, 120.9, 
61.54, 32.18, 473.93, 302.25, 0, 8.48, 3140.18, 0, 0, 6.83, 6.83, 
895.44, 895.44, 24.11, 24.11, 32.18, 32.18, 236.96, 236.96, 11.96, 
11.96, 80.08, 80.08, 3140.18, 3140.18, 163.62, 163.62, 236.96, 
236.96, 216.01, 216.01)), .Names = c("X", "AllwdAmt"), row.names = c(1137L, 
1138L, 1139L, 1140L, 1141L, 1142L, 1143L, 1144L, 1145L, 1146L, 
1147L, 1945L, 1946L, 1947L, 1948L, 1949L, 1950L, 1951L, 1952L, 
1953L, 1954L, 1955L, 1956L, 1957L, 1958L, 1959L, 2265L, 2266L, 
2267L, 2268L, 2269L, 2270L, 2271L, 2272L, 2273L, 2274L, 2275L, 
2276L, 2277L, 2278L, 2279L, 2280L, 2281L, 2282L, 2283L, 2284L, 
2285L, 2286L, 2287L, 2288L), class = "data.frame")

# get class intervals. This shows where the breaks should be.
library(classInt)
classIntervals(df$AllwdAmt, n = 10, style = 'jenks')

# Output:
style: jenks
   [0,140.48]   (140.48,396.26]   (396.26,799.55]  (799.55,1338.18] (1338.18,1864.02]        
         1109               423               111                97                58                 
  (1864.02,2586]    (2586,3451.35] (3451.35,5049.74] 
  12                44                20  
  (5049.74,6342.5] (6342.5,10407.88] 
           33                 2 

# A very inefficient way to assign breaks, based on output from above classIntervals:
df$AllwdBin <- ifelse(df$AllwdAmt <= 140.48,1,
                          ifelse(df$AllwdAmt > 140.48 & df$AllwdAmt <= 396.26,2,
                                 ifelse(df$AllwdAmt > 396.26 & df$AllwdAmt <= 799.55,3…

# The start of my code & associated error for automatically assigning a bin with breaks     
# coming from Jenks classification:

df$AllwdBin <- cut(df$AllwdAmt, breaks = classIntervals(df$AllwdAmt, n = 10, style =     
'jenks'), labels = c(as.character(1:10)))

# Output
Error in is.factor(x) : (list) object cannot be coerced to type 'double'

I understand the above error relates to the fact that the output of classintervals is producing a list of values, but how to turn that list into meaningful breaks?

回答1:

Look at the names of the output of classIntervals() like this:

foo <- classIntervals(df$AllwdAmt, n=10, style='jenks')
names(foo)

This will tell you that foo has two entries, varand brks.

You need to use the $brks component of this output:

cut(df$AllwdAmt, breaks = foo$brks, labels=as.character(1:10))

来源：https://stackoverflow.com/questions/24811952/create-binned-variable-from-results-of-class-interval-determination

标签

intervals

continuous

binning