A Prototype of Monotonic Binning Algorithm with R

May 4, 2013
By

(This article was first published on Yet Another Blog in Statistical Computing » S+/R, and kindly contributed to R-bloggers)

I’ve been asked many time if I have a piece of R code implementing the monotonic binning algorithm, similar to the one that I developed with SAS (http://statcompute.wordpress.com/2012/06/10/a-sas-macro-implementing-monotonic-woe-transformation-in-scorecard-development) and with Python (http://statcompute.wordpress.com/2012/12/08/monotonic-binning-with-python). Today, I finally had time to draft a quick prototype with 20 lines of R code, which is however barely useable without the further polish. But it is still a little surprising to me how efficient it can be to use R in algorithm prototyping, much sleeker than SAS macro.

library(sas7bdat)
library(Hmisc)

bin <- function(x, y){
  n <- min(50, length(unique(x)))
  repeat {
    n   <- n - 1
    d1  <- data.frame(x, y, bin = cut2(x, g = n)) 
    d2  <- aggregate(d1[-3], d1[3], mean)
    cor <- cor(d2[-1], method = "spearman")
    if(abs(cor[1, 2]) == 1) break
  }
  d2[2] <- NULL
  colnames(d2) <- c('LEVEL', 'RATE')
  head <- paste(toupper(substitute(y)), " RATE by ", toupper(substitute(x)), sep = '')
  cat("+-", rep("-", nchar(head)), "-+\n", sep = '')
  cat("| ", head, ' |\n', sep = '')
  cat("+-", rep("-", nchar(head)), "-+\n", sep = '')
  print(d2)
  cat("\n")
}

data <- read.sas7bdat("C:\\Users\\liuwensui\\Downloads\\accepts.sas7bdat")
attach(data)

bin(bureau_score, bad)
bin(age_oldest_tr, bad)
bin(tot_income, bad)
bin(tot_tr, bad)

R output:

+--------------------------+
| BAD RATE by BUREAU_SCORE |
+--------------------------+
       LEVEL       RATE
1  [443,618) 0.44639376
2  [618,643) 0.38446602
3  [643,658) 0.31835938
4  [658,673) 0.23819302
5  [673,686) 0.19838057
6  [686,702) 0.17850288
7  [702,715) 0.14168378
8  [715,731) 0.09815951
9  [731,752) 0.07212476
10 [752,776) 0.05487805
11 [776,848] 0.02605210

+---------------------------+
| BAD RATE by AGE_OLDEST_TR |
+---------------------------+
       LEVEL       RATE
1  [  1, 34) 0.33333333
2  [ 34, 62) 0.30560928
3  [ 62, 87) 0.25145068
4  [ 87,113) 0.23346304
5  [113,130) 0.21616162
6  [130,149) 0.20036101
7  [149,168) 0.19361702
8  [168,198) 0.15530303
9  [198,245) 0.11111111
10 [245,308) 0.10700389
11 [308,588] 0.08730159

+------------------------+
| BAD RATE by TOT_INCOME |
+------------------------+
           LEVEL      RATE
1 [   0,   2570) 0.2498715
2 [2570,   4510) 0.2034068
3 [4510,8147167] 0.1602327

+--------------------+
| BAD RATE by TOT_TR |
+--------------------+
    LEVEL      RATE
1 [ 0,12) 0.2672370
2 [12,22) 0.1827676
3 [22,77] 0.1422764

To leave a comment for the author, please follow the link and comment on their blog: Yet Another Blog in Statistical Computing » S+/R.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.

Sponsors

Mango solutions



plotly webpage

dominolab webpage



Zero Inflated Models and Generalized Linear Mixed Models with R

Quantide: statistical consulting and training

datasociety

http://www.eoda.de





ODSC

ODSC

CRC R books series





Six Sigma Online Training









Contact us if you wish to help support R-bloggers, and place your banner here.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)