Question

我有一个带有ID和起始日期的数据表，每个唯一ID代表一行。我必须使用变量'COUNT'（实际上是几个月内orig_date和close_date之间的间隔），并按顺序将ORIG_DATE复制到DATE字段中，如图所示。我尝试的代码只采用'COUNT'的第一个值（在这种情况下是3）并顺序复制ORIG_DATE。我对不同的ID有不同的COUNT。如何为每个唯一ID使用相应的COUNT，并将ORIG_DATE复制到另一个名为DATE

的列中

test.data

ID  COUNT   SCORE   VALUE   ORIG_DATE   CLOSE_DATE  
10748   3   750 450231  2015-03-01  2015-06-01      
10845   4   680 590231  2015-01-01  2015-05-01      
21758   7   760 650839  2014-11-01  2015-06-01   

test.panel <- test.data[rep(sequence(nrow(test.data)),COUNT)]

test.panel$DATE <- ymd(test.panel$ORIG_DATE)+ months(1:test.panel$COUNT)

以下是我正在尝试创建的数据表的结构

 ID COUNT   SCORE   VALUE   ORIG_DATE   DATE
10748   3   750 450231  2015-03-01  2015-03-01
10748   3   750 450231  2015-03-01  2015-04-01
10748   3   750 450231  2015-03-01  2015-05-01
10748   3   750 450231  2015-03-01  2015-06-01
10845   4   680 590231  2015-01-01  2015-01-01
10845   4   680 590231  2015-01-01  2015-02-01
10845   4   680 590231  2015-01-01  2015-03-01
10845   4   680 590231  2015-01-01  2015-04-01
10845   4   680 590231  2015-01-01  2015-05-01
21758   7   760 650839  2014-11-01  2014-11-01
21758   7   760 650839  2014-11-01  2014-12-01
21758   7   760 650839  2014-11-01  2015-01-01
21758   7   760 650839  2014-11-01  2015-02-01
..........................................................
..........................................................

Answer 1

It is actually simple to do this with data.table. Recreating your sample data:

test.data <- read.table( text = "
ID  COUNT   SCORE   VALUE   ORIG_DATE   CLOSE_DATE  
10748   3   750 450231  2015-03-01  2015-06-01      
10845   4   680 590231  2015-01-01  2015-05-01      
21758   7   760 650839  2014-11-01  2015-06-01", 
                 header = TRUE,
                 stringsAsFactors = FALSE,
                 colClasses = c("integer", "integer", "integer","integer", "Date", "Date") )
str(df)

Now doing what you want in data.table:

library(data.table)
test.data <- data.table(test.data)
test.data[ , list(CLOSE_DATE = seq(ORIG_DATE, CLOSE_DATE, by = "month")), 
   by = c("ID", "COUNT", "SCORE", "VALUE", "ORIG_DATE")]

       ID COUNT SCORE  VALUE  ORIG_DATE CLOSE_DATE
 1: 10748     3   750 450231 2015-03-01 2015-03-01
 2: 10748     3   750 450231 2015-03-01 2015-04-01
 3: 10748     3   750 450231 2015-03-01 2015-05-01
 4: 10748     3   750 450231 2015-03-01 2015-06-01
 5: 10845     4   680 590231 2015-01-01 2015-01-01
 6: 10845     4   680 590231 2015-01-01 2015-02-01
 7: 10845     4   680 590231 2015-01-01 2015-03-01
 8: 10845     4   680 590231 2015-01-01 2015-04-01
 9: 10845     4   680 590231 2015-01-01 2015-05-01
10: 21758     7   760 650839 2014-11-01 2014-11-01
11: 21758     7   760 650839 2014-11-01 2014-12-01
12: 21758     7   760 650839 2014-11-01 2015-01-01
13: 21758     7   760 650839 2014-11-01 2015-02-01
14: 21758     7   760 650839 2014-11-01 2015-03-01
15: 21758     7   760 650839 2014-11-01 2015-04-01
16: 21758     7   760 650839 2014-11-01 2015-05-01
17: 21758     7   760 650839 2014-11-01 2015-06-01

按R数据表中的ID顺序为每条记录创建日期

1 个答案: