Hi
Although I posted this in stackoverflow yesterday, I am asking here to get
helps as soon as quickly.
I need help make code for mocking sampling environment. Here is my code
below:
First, I generated mock units with 1000 groups of 100 units. Each row is
considered as independent sample space.
unit <- 100 # Total units
bad.unit.rate <- .05 # Proportion of bad units
bad.unit.num <- ceiling(unit*bad.unit.rate) # Bad units
n.sim=1000
unit.group <- matrix(0, nrow=n.sim, ncol=unit)for(i in 1:n.sim){
unit.group[i, ] <- sample(rep(0:1, c(unit-bad.unit.num, bad.unit.num)))}
dim(unit.group)
It gives 1000 by 100 groups
ss <- 44 # Selected sample size
44 out of 100 units will be selected and decision (pass or reject) will be
made based on sampling.
This below is decision code:
intercept <- rep(0, nrow(unit.group))
decision <- rep(0, nrow(unit.group))
set.seed(2017)for(i in 1:nrow(unit.group)){
selected.unit <- sample(1:unit, ss)
intercept[i] <- sum(unit.group[i,][selected.unit])
decision[i] <- ifelse(intercept[i]==0, 'pass', 'reject')
result <- cbind(intercept, decision)
result}
dim(result)
head(result, 30)
> head(result, 30)
intercept decision
[1,] "1" "reject"
[2,] "2" "reject"
[3,] "3" "reject"
[4,] "0" "pass"
[5,] "3" "reject"
[6,] "2" "reject"
[7,] "3" "reject"
[8,] "5" "reject"
[9,] "3" "reject"
[10,] "1" "reject"
[11,] "1" "reject"
[12,] "2" "reject"
[13,] "2" "reject"
[14,] "0" "pass"
[15,] "3" "reject"
[16,] "3" "reject"
[17,] "2" "reject"
[18,] "2" "reject"
[19,] "1" "reject"
[20,] "1" "reject"
[21,] "2" "reject"
[22,] "2" "reject"
I was able to make a decision for each 1000 rows based on sampling as above.
Now, I want to make code for "second" decision option as follows. Assuming
the row number is in order of time or sequence, if 'intercept' value is 0
or 'decision' is 'pass' in the row 4 above, I want to skip any decision
next following 5 (or else) and to label as 'skip', not 'reject'. In the
example above, rows from 5 to 9 will be 'skip' than 'reject'. Also, rows
from 15 to 19 should be 'skip' instead of 'reject'. Although I tried to
make preliminary code with my post, I have no idea where to start. Could
anyone help me to make code? Any feedback will be greatly appreciated.
Thank you very much in advance!!!
Steve
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
