汇总当年3个月的数据，然后转置R中一列的唯一值

Question

I have a column that represents a date in a dataframe. I simply want to create a new dataframe that summarizes amount coulmn at ID for 3months (20140831, 20140930,20141031) using the date column. Then transposing the Brand Column with the summarized Amount value for each. What is the best approach?

the data set is as below.

  ID         date       Brand     Amount
 1001   20141031    UNIBIC          9.8
 1001   20140930    UNIBIC          1.023
 1002   20140831    CITRIZINE       2.019
 1002   20140930    CITRIZINE       2.015
 1002   20141031    CITRIZINE       1.002
 1003   20140831    CHOCO           4.22
 1004   20140930    SOLOSTAR        1.007
 1004   20141030    SOLOSTAR        1.008
 1005   20140930    DOLO            1.025

I would like to have the Output as below

  ID           UNIBIC     CITRIZINE   CHOCO   SOLOSTAR      DOLO  
  1001        5.4115                
  1002                  1.678
  1003                               4.22        
  1004                                        1.039
  1005                                                    1.025

Any help provided from your end will be greatly appreciated

Answer 1

You can try

library(reshape2)
df$Brand <- factor(df$Brand, levels=unique(df$Brand))
dcast(df, ID~Brand, value.var='Amount', mean)
#   ID UNIBIC CITRIZINE CHOCO SOLOSTAR  DOLO
#1 1001 5.4115       NaN   NaN      NaN   NaN
#2 1002    NaN  1.678667   NaN      NaN   NaN
#3 1003    NaN       NaN  4.22      NaN   NaN
#4 1004    NaN       NaN   NaN   1.0075   NaN
#5 1005    NaN       NaN   NaN      NaN 1.025

Or may be you can try dcast.data.table which would be faster

library(data.table) 
dcast.data.table(setDT(df), ID~Brand, value.var='Amount', mean)

Or using dplyr/tidyr

library(dplyr)
library(tidyr)

df %>%
   group_by(ID, Brand) %>% 
   summarise(Amount=mean(Amount)) %>%
   ungroup() %>%
   spread(Brand, Amount)

Update

If you need only Aug , Sep , and Oct , you could subset the dataset before transforming.

df1 <-  df[as.numeric(substr(df$date, 5,6)) %in% 8:10,]
dcast(df1, ID~Brand, value.var='Amount', mean)

data

 df <- structure(list(ID = c(1001L, 1001L, 1002L, 1002L, 1002L, 1003L, 
 1004L, 1004L, 1005L), date = c(20141031L, 20140930L, 20130831L, 
 20140930L, 20141031L, 20130831L, 20130930L, 20131030L, 20140930L
 ), Brand = c("UNIBIC", "UNIBIC", "CITRIZINE", "CITRIZINE", "CITRIZINE", 
 "CHOCO", "SOLOSTAR", "SOLOSTAR", "DOLO"), Amount = c(9.8, 1.023, 
 2.019, 2.015, 1.002, 4.22, 1.007, 1.008, 1.025)), .Names = c("ID", 
 "date", "Brand", "Amount"), class = "data.frame", row.names = c(NA, -9L))