Bind Tables into one data frame - r

Question

I have searched a couple of options, generally trying out various combinations on cbind to accomplish this. Essentially I would like to create a data frame that combines different pivot tables. into one data frame in order to export to csv/excel. Is there a better way to accomplish this?
EDIT: Essentially I am trying to learn the basics of creating a function that can wrap around multiple different pivot tables to create a data frame ready for export that will serve as a template to ad hoc reporting. The problem I am having is that the cbind product takes object B, which as a standalone will be a table with the dates as columns, and forces it into a long table, where the dates are transposed into rows.
dataframe:
State FacilityName Date
NY Loew June 2014
NY Loew June 2014
CA Sunrise May 2014
CA May 2014
code:
volume <- function() {
df$missing = ifelse(is.na(df$FacilityName), "Missing", df$FacilityName)
df = subset(df, df$missing == "Missing")
x <- function(){
a <- as.data.frame(table(df$FacilityName))
b <- table(df$FacilityName, df$date)
cbind(a, b[,1], b[2])
}
}

Answer 1

When you give a factor to the table function, it uses the levels of the factor to build the table. So there's a nice way to obtain what you want by adding "Missing" to the levels of "FacilityName".
# loading data
ec <- read.csv(text=
'State, FacilityName, Date
NY,Loew,June 2014
NY,Loew,June 2014
CA,Sunrise,May 2014
CA,NA,May 2014', )
# Adding Missing to the possible levels of FacilityName
# note that we add it in front
new.levels <- c("Missing", levels(ec$FacilityName))
ec$FacilityName <- factor(ec$FacilityName, levels=new.levels)
# And replacing NAs by the new level "Missing"
ec$FacilityName[is.na(ec$FacilityName)] <- "Missing"
# the previous line would not have worked
# if we had not added "Missing" explicitly to the levels
# table() uses the levels to generate the table
# the levels are displayed in order
# now there's a level "Missing" in first position
t <- table(ec$FacilityName, ec$Date)
You get:
> t
June 2014 May 2014
Missing 0 1
Loew 2 0
Sunrise 0 1
You can add the total line like this (I don't think your code with nrow do what you say it does)
# adding total line
rbind(t, TOTAL=colSums(as.matrix(t)))
June 2014 May 2014
Missing 0 1
Loew 2 0
Sunrise 0 1
TOTAL 2 2
At this point you have a matrix so you may want to pass it to as.data.frame.
This can be easily implemented into a separate function if you want to. No need to bind several tables after all :)

Answer 2

Ok, so it seems I was trying to be cool and use a function to wrap everything in the hopes that it would be the beginning of learning to write flexible code. But, I did it the long way and ended up getting the result I wanted. While I will post the code that worked below, I am very interested in someone pointing me towards a better way to approach these kinds of problems, in order to learn better coding.
# Label the empty cells as Missing
ec$missing = ifelse(is.na(ec$FacilityName), "Missing", ec$FacilityName)
# Subset the dataframe to just missing values
df = subset(ec, ec$missing == "Missing")
# Create table that is a row of frequency by month for missing values
a <- table(df$missing, df$date)
# Reload dataframe to exclude Missing values
df = subset(ec, ec$missing != "Missing")
# Create table that shows frequency of observations for each facility by Month
b <- table(df$FacilityName, df$date)
# Create a Total row that can go at the bottom of the final data frame
Total <- nrow(ec)
# Bind all three objects
rbind(a,b,Total)
Here is an example of the final product I was looking for:
May2014 June2014
Missing 2 0
Sunrise 0 0
Loew 1 2
Total 3 2