CODEWITHSUNDEEP
rdataframematrix
h7ral.shah
h7ral.shah

Reputation: 13

How to add a matrix to a data.frame?

below is some dummy data.

Suppose I have a dataframe

df = data.frame(source = c("X1", "X2", "X3", "X4", "X5", "X6", "X7", "X8", "X9", "X10",
                       "X11", "X12", "X13", "X14", "X15", "X16", "X17", "X18", "X19", "X110"),
                Destination = c("X3","X5","X17", "X20", "X20","X1", "X2", "X3", "X7", "X10", 
                                "X13","X15","X7", "X1", "X20","X17", "X2", "X3", "X7", "X10"),
                weight = seq(1,1.95,by=0.05))

I then have some odds ratios for Destinations X1:X3 and there respective standard deviations and I want to randomly sample 10 times from each odds ratio and its corresponding standard deviation

OR_dat <- c(1.55,1.39,1.77)

sds <- c(0.2925175, 0.4775346, 0.1603566)
n <- 10

normv <- function( n , mean , sd ){
  out <- rnorm( n*length(mean) , mean = mean , sd = sd )
  return( matrix( out , ncol = n , byrow = FALSE ))
}

RR_neighbour_1 <- data.frame(t(normv(n, OR_dat , sds )))
colnames(RR_neighbour_1) <- c("X1",  "X2",  "X3")

What I am really looking for is merging the matrix into the data.frame by looking at the value in the column titled "Destination", matching it up with the column name of the matrix titled RR_neighbour_1 and then creating additional rows to input the distribution. The output should then end up like the following:

enter image description here

Upvotes: 1

Views: 1018

Answers (2)

Jon
Jon

Reputation: 455

One possibility: if you're happy to use the dplyr package then it includes SQL-style join functions. You probably want the left_join function from that package, which lets you map columns using the by parameter. That's an easy way to join two table-like structures.

Upvotes: 0

jkd
jkd

Reputation: 1664

What you actually want to do is to merge two data.frames by Destination. So you first need to get your second data.frame (RR_neighbour_1) into long format (so the same format as the first one, where the different destinations are rows not columns). Then you can simply merge the data.frames with the merge function. Argument all=T will ensure that the rows for the repeated destinations will be added.

RR_neighbour_1 <- reshape(RR_neighbour_1,dir="long",varying = list(1:3),
                          timevar = "Destination",
                          times = colnames(RR_neighbour_1),
                          v.names = "RR_neighbour_1")

merge(df, RR_neighbour_1[,-3], all=T)

Upvotes: 1

Related Questions