On May 20, 2010, at 6:38 AM, Mohan L wrote:
Dear All,
I have data some thisng like this :
data <- read.csv(file='ipsample.csv',sep=',' , header=TRUE)
data
State Jan Feb Mar Apr May Jun
1 AAA 1 1 0 2 2 0
2 BBB 1298 1195 1212 1244 1158 845
3 CCC 0 0 0 1 2 1
4 DDD 5 11 17 15 10 9
5 EEE 18 28 27 23 23 16
6 FFF 68 152 184 135 111 86
from this data frame, I took "Jan" as base and calculating weightage
like
this :
basemonth.sum <- sum(data[[2]])
basemonth.sum
[1] 1390
basemonth.data <- data[[2]]
basemonth.data
[1] 1 1298 0 5 18 68
weightage <- basemonth.data / basemonth.sum
weightage
[1] 0.0007194245 0.9338129496 0.0000000000 0.0035971223 0.0129496403
[6] 0.0489208633
The above is the weightage for base month "Jan". Now I need to
calculate
weighted states data. What I need to do is :
(((Feb[i]-Jan[1])*weightage)+Jan[1]) for all column. The "Jan"
column is
fixed. I need to do the calculation in all the column Feb, Mar etc...
data[, 3:7]*(
data[ , 2]/sum(data[ , 2]) )
Gives the reweighted estimates. You could easily cbind them to data[ ,
1:2]
State Jan Feb
Mar
1 AAA 1 (((Feb[1]-Jan[1])*weightage[1])+Jan[1])
(((Mar[1]-Jan[1])*weightage[1])+Jan[1])
2 BBB 1298 (((Feb[2]-Jan[2])*weightage[2])+Jan[2])
(((Mar[2]-Jan[1])*weightage[2])+Ja[1])
3 CCC 0
4 DDD 5
5 EEE 18
6 FFF 68
I am struggling with this . I have framed a logic using for loop.
But it
seems me very bad logic. Any help will be greatly appreciated.
Thanks & Rg
Mohan L
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
David Winsemius, MD
West Hartford, CT
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
