Constructing the transition pair using loop function and paste()

newtrans <- lapply(split(mydata, mydata$agents), function(.agent){

newtrans

I got it! Thank you so much Jim!

On 20 October 2011 15:06, jim holtman <jholtman at gmail.com> wrote:

try this: ?You were redfining 'transition' within the loop

x <-

c('A','D','F','H','N','A','C','H','F','D','F','F','H','K','G','D','D','B','N','K','O','V','S','S','F','H','J','U','K','T','H','S','R','G','S','R','R','G','D','B','G','F','G','N','H','H','K','L','B','X','C','V','S','F','X','G','T','H','H','H','R','A','E','D','C')
y <- c(rep(0293,8), rep(0498,6), rep(6847,10), rep(6209,4), rep(8562,13),
rep(4596,6), rep(2857,2), rep(6178,3), rep(6018,5), rep(5629,4),
rep(7535,4))
mydata <- as.data.frame(cbind(x,y))
names(mydata) <- c('actions', 'agents')
mydata

transition <- vector('list', length(unique(mydata$agents))) ? ? ? ?#
create a list to hold the outputs

for (i in 1:length(unique(mydata$agents))){ ? ? ? ? # decompose the
data frame by agents
? ?agent.i <- mydata[mydata$agents == (unique(mydata$agents))[i], ]
? ?transit.i <- c()
? ?for (j in 1:length(agent.i$actions)-1){
? ? ? ?transit.i[j] <- paste(agent.i$actions[j],
agent.i$actions[j+1], sep = '')
? ?}
? ?# for each subset of each agent, perform the 'pairing'
? ?transition[[i]] <- transit.i
}
transition

On Thu, Oct 20, 2011 at 8:52 AM, Sally Zhen <saliluna at gmail.com> wrote:

Hi all,


I'd like to thank those who helped me with my previous loop function
question with agents/events. I have solved the problem with the advice
from
this community.

I have now moved on to the next step, which requires me to find all the
transition pair within an event. A sample data and the R commands I've
written are as follow:

x <-

c('A','D','F','H','N','A','C','H','F','D','F','F','H','K','G','D','D','B','N','K','O','V','S','S','F','H','J','U','K','T','H','S','R','G','S','R','R','G','D','B','G','F','G','N','H','H','K','L','B','X','C','V','S','F','X','G','T','H','H','H','R','A','E','D','C')
y <- c(rep(0293,8), rep(0498,6), rep(6847,10), rep(6209,4),
rep(8562,13),
rep(4596,6), rep(2857,2), rep(6178,3), rep(6018,5), rep(5629,4),
rep(7535,4))
mydata <- as.data.frame(cbind(x,y))
names(mydata) <- c('actions', 'agents')
mydata


for (i in 1:length(unique(mydata$agents))){ ? ? ? ? # decompose the data
frame by agents
agent.i <- mydata[mydata$agents == (unique(mydata$agents))[i], ]
transition <- vector('list', length(unique(mydata$agents))) ? ? ? ?#
create
a list to hold the outputs
transit.i <- c()
for (j in 1:length(agent.i$actions)-1){
transit.i[j] <- paste(agent.i$actions[j], agent.i$actions[j+1], sep =
'')}
? # for each subset of each agent, perform the 'pairing'
transition[[i]] <- transit.i}
transition


The actions are ordered, so what I need to do is just to paste each
action
to the next one to form a pair.
My attempt only produced the desired result for the last agent:

[[1]]
NULL

[[2]]
NULL

[[3]]
NULL

[[4]]
NULL

[[5]]
NULL

[[6]]
NULL

[[7]]
NULL

[[8]]
NULL

[[9]]
NULL

[[10]]
NULL

[[11]]
[1] "AE" "ED" "DC"


Is there any way to avoid using 2-level loop function, although to me
it's
the most intuitive method? Any pointer would be greatly appreciated!

Regards,
Sally

? ? ? ?[[alternative HTML version deleted]]

--
Jim Holtman
Data Munger Guru

What is the problem that you are trying to solve?