The question was phrased weirdly, but I'll try my absolute best to explain it thoroughly.
I have a data frame that was constructed from a time series object. The time series object was constructed like so:
std<- ts(baser, frequency=12, start=c(1987,1))
And the data.frame that was built from this previous ts object was built like so:
meses<- c("Enero","Febrero","Marzo","Abril","Mayo","Junio","Julio","Agosto","Septiembre","Octubre","Noviembre","Diciembre")
td<- ts2df(std,mon1=1)
names(td)<-meses
Note that the function ts2df is not a part of core R, but taken from the package 'wq' This obviously forces me to create a vector of names that will be coerced into the name of the data frame (here, in spanish).
The resulting data.frame is naturally large, with 12 columns (with names), and enough rows for every eyar from 1987 to 2013.
The summary function for this data.frame is expressed in the same format (same number of columns), with 6 rows.
I need to export this summary with LaTeX formatting, but the table is (naturally) too long. The package xtable allows to do this, but still keeps the table with 12 columns wide. The most symmetrical would be to get the summary with 6 columns: The months from January to June, the summary statistics below them. And below those, again the remaining months (July to December), with their respective statistics underneath.
I am willing and able to modify any steps in order to get what I need, if the solution is easier in any of them.