Group results under data names with sapply function and plot in R - r

Question

Let's say I have a data which involves 3 separate data. Here is my data;
data<-structure(list(x = structure(list(value = c(2L, 4L, 5L, 6L, 9L,
4L, 3L, 2L, 10L, 6L)), .Names = "value", class = "data.frame", row.names = c(NA,
-10L)), y = structure(list(value = c(2, 2.1, 4, 3, 0, 1.2, 4.2,
3, 4, 9)), .Names = "value", class = "data.frame", row.names = c(NA,
-10L)), z = structure(list(value = c(1, 2, 7, 6, 0.3, 5.4, 4,
3, 6, 7)), .Names = "value", class = "data.frame", row.names = c(NA,
-10L))), .Names = c("x", "y", "z"))
And here is my sample function;
sam<-function(x) {
ex<-c(3,2,4,5,2)
z<-data.frame(x)
y<-as.matrix(sapply(z, as.numeric))
h<-lapply(c(2,5,10), function(xx) tapply(y, as.integer(gl(nrow(x), xx, nrow(x)) ), FUN = sum))
names(h)<-c("min2", "min5", "min10")
min2<-h[[1]]
pdf("plots.pdf")
plot(min2, ex, main="min. compare",
xlab="Historical Values ", ylab="Disaggregated Values", pch=19, col = "blue")
dev.off()
return(h)
}
In the function, I am aggregating values as shown. And then plotting min2 with ex data.
With the code below, I tried to use the function for all data like;
v1<-sapply(data, sam)
But I can not see calculation's name as min2 min5 min10 in result list. And also results are coming complexly, not under the x, y and z
I desire these two;
1) Grouping results under each data name. Like;
[x] [y] [z]
min2 min2 min2
min5 min5 min5
min10 min10 min10
2) Plotting the desired ones for all x, y and z as mentioned above. And export three plots to one pdf or separately.

Answer 1

To get the output, like #JonnyPhelps suggested, use lapply instead of sapply. To make the plots and get correlation you need to alter the function:
sam<-function(x) {
ex<-c(3,2,4,5,2)
z<-data.frame(x)
y<-as.matrix(sapply(z, as.numeric))
h<-lapply(c(2,5,10), function(xx) tapply(y, as.integer(gl(nrow(x), xx, nrow(x)) ), FUN = sum))
names(h)<-c("min2", "min5", "min10")
min2<-h[[1]]
plot(min2, ex, main="min. compare",
xlab="Historical Values ", ylab="Disaggregated Values",
pch=19, col = "blue")
COR = cor.test(min2,ex)
LABEL = paste("cor=",signif(COR$estimate,3),"\np=",signif(COR$p.value,3))
mtext(LABEL,side=3,padj=2)
return(h)
}
The correlation is calculated and you use mtext to place it at the top of the plot. You can play around with padj and adj to get the text where you need.
In your previous function, you called the plot in the function, this overwrites the file with every iteration. To plot all on a pdf, you need to do:
pdf("plots.pdf")
v1<-lapply(data, sam)
dev.off()
Or if you want them on the same page:
pdf("plots.pdf",width=8,height=4)
par(mfrow=c(1,3))
v1<-lapply(data, sam)
dev.off()