I have some data that looks-like this:
id1 id2 id3 id4 amt
2021-06-15 2021-06-30 2021-07-15 2021-07-30 10
2021-06-15 2021-06-30 2021-07-15 2021-07-30 20
2021-07-15 2021-06-30 2021-07-15 2021-07-30 30
2021-07-30 2021-06-30 2021-07-15 2021-07-30 10
and I create a new column with
dat$month <- as.Date(cut(dat$id1, breaks="month"))
in order to sum the amounts by month so that I can cbind the amount like this:
bymonth <- aggregate(cbind(amt)~month, data=dat,FUN=sum)
I want to sum by month for each id#, and then cbind an aggregate ~month for each id#. Do I need to create a month column for each id - like month-id1, monthid2, etc., and then cbind a bymonth-id1m, bymonth-id2, etc.? I eventually create a time series for each id#, so I think it would work to create these one-by-one, but it doesn't feel very elegant.
question from:
https://stackoverflow.com/questions/66056120/cut-and-aggregate-multiple-columns 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…