[R] Arrange data

[Rasmus Liland]

On 2020-08-03 21:11 +1000, Jim Lemon wrote:
> On Mon, Aug 3, 2020 at 8:52 PM Md. Moyazzem Hossain <hossainmm using juniv.edu> wrote:
> >
> > Hi,
> >
> > I have a dataset having monthly 
> > observations (from January to 
> > December) over a period of time like 
> > (2000 to 2018). Now, I am trying to 
> > take an average the value from 
> > January to July of each year.
> >
> > The data looks like
> > Year    Month  Value
> > 2000    1         25
> > 2000    2         28
> > 2000    3         22
> > ....    ......      .....
> > 2000    12       26
> > 2001     1       27
> > .......         ........
> > 2018    11       30
> > 20118   12      29
> >
> > Can someone help me in this regard? 
> >
> > Many thanks in advance.
> 
> Hi Md,
> One way is to form a subset of your 
> data, then calculate the means by 
> year:
> 
> # assume your data is named mddat
> mddat2<-mddat[mddat$month < 7,]
> jan2jun<-by(mddat2$value,mddat2$year,mean)
> 
> Jim

Hi Md,

you can also define the period in a new 
column, and use aggregate like this:

	Md <- structure(list(
	Year = c(2000L, 2000L, 2000L, 
	2000L, 2001L, 2018L, 2018L), 
	Month = c(1L, 2L, 3L, 12L, 1L,
	11L, 12L), 
	Value = c(25L, 28L, 22L, 26L,
	27L, 30L, 29L)), 
	class = "data.frame", 
	row.names = c(NA, -7L))
	
	Md[Md$Month %in%
	        1:6,"Period"] <- "first six months of the year"
	Md[Md$Month %in% 7:12,"Period"] <- "last six months of the year"
	
	aggregate(
	  formula=Value~Year+Period,
	  data=Md,
	  FUN=mean)

Rasmus