Row comparison in R - r

I have the below data frame,
R_Number A
1 0
2 15
3 10
4 11
5 12
6 18
7 19
8 15
9 17
10 11
Now I need to create another column B where the comparison of the values in A will be computed. The condition is that the comparion is not between two consecutive row, i.e Row number 1 is compared with Row number 4, like wise Row number 2 is compared with Row number 5 and this iteration continues till the end of the data . Condition for comparision result is:
if (A[1]>=15 && A[4] <= 12) {
B == 1
}
else if (A[1]<=0 && A[4]>= 10) {
B== 2
}
else {
B== 0
}
When it comes to Row number 8 and Row number 9these rows will not have next 4th row to compare with hence the value should be 0
Also, the comparision result of Row 1 and 4 is printed in Row number 1 similarly comparision result of Row 2 and 5 is printed in Row number 2
So the resulting dataframe should be as shown below
R_Number A B
1 0 2
2 15 1
3 10 0
4 11 0
5 12 0
6 18 0
7 19 1
8 15 0
9 17 0
10 11 0

According to #nicola comment, I tried to solve your problem as well.
I recreated your initial data frame:
df <- data.frame(R_Number = c(1:10), A = c(0,15,10,11,12,18,19,15,17,11), B = 0)
So I used an if statement inside a cycle for:
for (i in 1:(length(df$A)-3)) {
if (df$A[i] >= 15 && df$A[i+3] <= 12) {
df$B[i] <- 1
} else if ((df$A[i] <= 0) && (df$A[i+3] >= 10)) {
df$B[i] <- 2
}
else {
df$B[i] <- 0
}
}
With last edit I solved the problem that came up when the length of data frame changed.
Now you have a generic solution!

First lagging the variable and then computing your new variable should work. Something like this:
library(Hmisc)
df <- data.frame(R_Number = c(1:10), A = c(0,15,10,11,12,18,19,15,17,11))
A_Lag<-Lag(df$A,-3)
df$B <- rowSums(cbind(df$A>=15 & A_Lag <= 12,(df$A<=0 & A_Lag>= 10)*2),na.rm= T)
df$B
I tried to avoid if statements. The Lag function can be found in the Hmisc package.
> df$B
[1] 2 1 0 0 0 0 1 0 0 0

Related

Find the number of previous elements in vector until current value was equal or exceeded

For every element in a vector, I want to calculate the number of previous elements until I reach an element that is equal to or exceeds the value of the current element.
This is my vector:
CPC= c(25382.6, 30044.9, 22664.4, 30752.3, 21896.9, 24173.1, 29777.9, 9021.1, 8175.1, 9005.8, 5314.2, 4274.1, 3759.1, 5043.1, 5080.9, 6444.4, 6097.6, 8004.2, 6004.7, 6468.9, 5104.7, 5985.5, 8343.7, 8582, 7494.3, 6088.9, 4372.7, 4298.6, 4553.2, 5742)
I have tried something like this, which is not working
ROC = NULL #Create vector to store values
for (i in seq(1:length(CPC))
while(CPC[i])<CPC[i-1]
ROC[i] <- ifelse((CPC[i] < CPC[i-1]),1,0)
The output I am looking for is a vector (ROC) of same lenght as the original one (CPC), so that every element corresponds to the element on the same place in the original vector. ROC[i] would for CPC[i] give the number of previous elements in the CPC vector until the value is equal to or bigger than CPC[i]. For some elements, there are no pervious elements that has a higher or equal value and this would still need to be saved in the output, for example as a NA.
It would look like this:
ROC
NA,NA,1,NA,1,2,3,1,1,2... etc
prs <- unlist(lapply(1:(length(CPC)),
function(x) {
less_or_eq <- CPC[1 : x] <= CPC[x]
if(all(less_or_eq))
return(0)
inds <- which(less_or_eq == 0)
return(x - inds[length(inds)])
} ))
prs
# 0 0 1 0 1 2 3 1 1 2 1 1 1 3 4 6 1 8 1 2 1 2 13
# 14 1 1 1 1 3 4
Using for loop
res <- numeric(length(CPC))
for (i in 1 : length(CPC)) {
less_or_eq <- CPC[1 : i] <= CPC[i]
if (all(less_or_eq)) {
res[i] <- 0
} else {
inds <- which(less_or_eq == 0)
res[i] <- (i - inds[length(inds)])
}
}
res
#[1] 0 0 1 0 1 2 3 1 1 2 1 1 1 3 4 6 1 8 1 2 1 2 13
#[24] 14 1 1 1 1 3 4
Your output requirement is not very clear. The main part is to find where the difference (diff) changes to from negative to positive (meaning that the previous value is less than or equal to the current one. We create the groups based on that and count the length of each group, i.e.
tapply(CPC, cumsum(c(TRUE, diff(CPC) >= 0)), length)
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
# 1 2 2 1 3 4 1 1 2 2 2 1 1 5 1 1
#or take the cumulative sum of the above,
cumsum(tapply(CPC, cumsum(c(TRUE, diff(CPC) >= 0)), length))
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
# 1 3 5 6 9 13 14 15 17 19 21 22 23 28 29 30

Increment call vector on itself in R - Finding counts in between values

I have a time series (or simply a vector) that is binary, returning 0 or 1's depending on some condition (generated with ifelse). I would like to be able to return the counts (in this case corresponds to time series, so days) in between the 1's.
I can do this very easily in Excel, by simply calling the Column I am trying to calculate and then adding the row above (if working with Ascending data, or calling row below if working with descending). See below
I tried doing something similar in R but I am getting an error.
DaysBetweenCondition1 = as.numeric(ifelse((Condition1 ==0 ),0,lag(DaysBetweenCondition1)+1))
Is there an easier way to do this besides making a function
Row# Date Condition1 DaysBetweenCondition1
1 5/2/2007 NA NA
2 5/3/2007 NA NA
3 5/4/2007 NA NA
4 5/5/2007 NA NA
5 5/6/2007 0 NA
6 5/7/2007 0 NA
7 5/8/2007 0 NA
8 5/9/2007 0 NA
9 5/10/2007 0 NA
10 5/11/2007 0 NA
11 5/12/2007 0 NA
12 5/13/2007 0 NA
13 5/14/2007 1 0
14 5/15/2007 0 1
15 5/16/2007 0 2
16 5/17/2007 0 3
17 5/18/2007 0 4
18 5/19/2007 0 5
19 5/20/2007 0 6
20 5/21/2007 0 7
21 5/22/2007 1 0
22 5/23/2007 0 1
23 5/24/2007 0 2
24 5/25/2007 0 3
25 5/26/2007 0 4
26 5/27/2007 1 0
27 5/28/2007 0 1
28 5/29/2007 0 2
29 5/30/2007 1 0
(fwiw, the Dates in this example are made up, in the real data I am using business days so a bit different, and I dont want to reference them, just put in for clarity)
This gets the counting done in one line. Borrowing PhiSeu's code and a line from How to reset cumsum at end of consecutive string and modifying it to count zeros:
# Example
df_date <- cbind.data.frame(c(1:20),
c(rep("18/08/2016",times=20)),
c(rep(NA,times=5),0,1,0,0,1,0,0,0,0,1,1,0,1,0,0)
,stringsAsFactors=FALSE)
colnames(df_date) <- c("Row#","Date","Condition1")
# add the new column with 0 as default value
DaysBetweenCondition1 <- c(rep(0,nrow(df_date)))
# bind column to dataframe
df_date <- cbind(df_date,DaysBetweenCondition1)
df_date$DaysBetweenCondition1<-sequence(rle(!df_date$Condition1)$lengths) * !df_date$Condition1
R is very good when working with rows that don't depend on each other. Therefore a lot of functions are vectorized. When working with functions that depend on the value of other rows it is not so easy.
At the moment I can only provide you with a solution using a loop. I assume there is a better solution without a loop.
# Example
df_date <- cbind.data.frame(c(1:20),
c(rep("18/08/2016",times=20)),
c(rep(NA,times=5),0,1,0,0,1,0,0,0,0,1,1,0,1,0,0)
,stringsAsFactors=FALSE)
colnames(df_date) <- c("Row#","Date","Condition1")
# add the new column with 0 as default value
DaysBetweenCondition1 <- c(rep(0,nrow(df_date)))
# bind column to dataframe
df_date <- cbind(df_date,DaysBetweenCondition1)
# loop over rows
for(i in 1:nrow(df_date)){
if(is.na(df_date$Condition1[i])) {
df_date$DaysBetweenCondition1[i] <- NA
} else if(df_date$Condition1[i]==0 & is.na(df_date$Condition1[i-1])) {
df_date$DaysBetweenCondition1[i] <- NA
} else if(df_date$Condition1[i]==0) {
df_date$DaysBetweenCondition1[i] <- df_date$DaysBetweenCondition1[i-1]+1
} else {
df_date$DaysBetweenCondition1[i] <- 0
}
}
Here's a solution that should be relatively fast
f0 = function(x) {
y = x # template for return value
isna = is.na(x) # used a couple of times
grp = cumsum(x[!isna]) # use '1' to mark start of each group
lag = lapply(tabulate(grp + 1), function(len) {
seq(0, length.out=len) # sequence from 0 to len-1
})
split(y[!isna], grp) <- lag # split y, set to lag element, unsplit
data.frame(x, y)
}
A faster version avoids the lapply() loop; it creates a vector along x (seq_along(x)) and an offset vector describing how the vector along x should be corrected based on the start value of the original vector
f1 = function(x0) {
y0 = x0
x = x0[!is.na(x0)]
y = seq_along(x)
offset = rep(c(1, y[x==1]), tabulate(cumsum(x) + 1))
y0[!is.na(y0)] = y - offset
data.frame(x0, y)
}
Walking through the first solution, here's some data
> set.seed(123)
> x = c(rep(NA, 5), rbinom(30, 1, .15))
> x
[1] NA NA NA NA NA 0 0 0 1 1 0 0 1 0 0 1 0 0 0 0 1 0 0 0 1
[26] 1 0 0 1 0 0 0 0 0 0
use cumsum() to figure out the group the non-NA data belong to
> isna = is.na(x)
> grp = cumsum(x[!isna])
> grp
[1] 0 0 0 1 2 2 2 3 3 3 4 4 4 4 4 5 5 5 5 6 7 7 7 8 8 8 8 8 8 8
use tabulate() to figure out the number of elements in each group, lapply() to generate the relevant sequences
> lag = lapply(tabulate(grp + 1), function(len) seq(0, length.out=len))
finally, create a vector to hold the result, and use spilt<- to update with the lag
> y = x
> split(y[!isna], grp) <- lag
> data.frame(x, y)
x y
1 NA NA
2 NA NA
3 NA NA
4 NA NA
5 NA NA
6 0 0
7 0 1
8 0 2
9 1 0
10 1 0
11 0 1
12 0 2
13 1 0
14 0 1
15 0 2
16 1 0
17 0 1
...
The key to the second solution is the calculation of the offset. The goal is to be able to 'correct' y = seq_along(x) by the value of y at the most recent 1 in x, kind of like 'fill down' in Excel. The starting values are c(1, y[x==1]) and each needs to be replicated by the number of elements in the group tabulate(cumsum(x) + 1).

R apply returning rows but want columns

I am trying to clean survey data where sometimes minutes information is entered in the hours field. The cleaning rules say to move the hours value to the minutes field if mins is missing or 0 and hours is 15 or 30 or 45 or 60.
Here is the function I wrote:
cleanHrMin <- function(x){
h = x[1]
m = x[2]
if ( !is.na(h) && (h==15 || h==30 || h==45 || h==60) && (m==0 || is.na(m)) )
{ return(c(0,h)) }
else
{ return(x) }
}
With test data:
df <- as.data.frame(cbind(hrs = c(1,15,0), mins = c(10,NA,15)))
I run the function as follows:
as.data.frame(apply(df,1,cleanHrMin))
and get output:
V1 V2 V3
1 1 0 0
2 10 15 15
But what I want is:
V1 V2
1 1 10
2 0 15
3 0 15
What am I doing wrong?
You're doing the correct apply, across rows. You simply need to transpose the output.
> as.data.frame(t(apply(df,1,cleanHrMin)))
V1 V2
1 1 10
2 0 15
3 0 15
This is because apply returns columns resulting from the function, whether your margin is over rows or over columns. So to go from rows to rows, you must transpose.

How to extract all rows between start signal and end signal?

I have the following df and I would like to extract all rows based on the following start and end signals.
Start signal : When status changes from 1 to 0
End signal : When status changes from 0 to -1.
df <- data.frame(time = rep(1:14), status = c(0,1,1,0,0,0,-1,0,1,0,0,0,-1,0))
time status
1 1 0
2 2 1
3 3 1
4 4 0
5 5 0
6 6 0
7 7 -1
8 8 0
9 9 1
10 10 0
11 11 0
12 12 0
13 13 -1
14 14 0
Desire:
time status
4 4 0
5 5 0
6 6 0
10 10 0
11 11 0
12 12 0
Here's a possible solution using the data.table package. I'm basically first grouping by status == 1 appearances and then checking per group if there was also a status == -1, if so, I'm sub-setting the group from the second incident until the -1 incident minus 1
library(data.table)
setDT(df)[, indx := cumsum(status == 1)]
df[, if(any(status == -1)) .SD[2:(which(status == -1) - 1)], by = indx]
# indx time status
# 1: 2 4 0
# 2: 2 5 0
# 3: 2 6 0
# 4: 3 10 0
# 5: 3 11 0
# 6: 3 12 0
We count start and end markers, then use those values and the cumulative-sum of (start - end) to filter rows. The (cumsum(start)-cumsum(end)>1) is a slight fiddle to avoid the cumulative counts being upset by row 2 which starts but doesn't end; otherwise row 14 would unwantedly get included.
require(dplyr)
df %>% mutate(start=(status==1), end=(status==-1)) %>%
filter(!start & !end & (cumsum(start)-cumsum(end)>1) ) %>%
select(-start, -end)
# time status
# 1 4 0
# 2 5 0
# 3 6 0
# 4 10 0
# 5 11 0
# 6 12 0
A little ugly, but you can always just loop over the values and keep a flag for determining whether the element should be kept or not.
keepers <- rep(FALSE, nrow(df))
flag <- FALSE
for(i in 1:(nrow(df)-1)) {
if(df$status[i] == 1 && df$status[i+1] == 0) {
flag <- TRUE
next # keep signal index false
}
if(df$status[i] == -1 && df$status[i+1] == 0) {
flag <- FALSE
next # keep signal index false
}
keepers[i] <- flag
}
keepers[nrow(df)] <- flag # Set the last element to final flag value
newdf <- df[keepers, ] # subset based on the T/F values determined
Do you have some more data (or can you gen some more data you know the outcome of) to see if this/these generalize?
Two similar approaches:
library(stringr)
df <- data.frame(time = rep(1:14), status = c(0,1,1,0,0,0,-1,0,1,0,0,0,-1,0))
dfr <- rle(df$status)
# first approach
find_seq_str <- function() {
str_locate_all(paste(gsub("-1", "x", dfr$values), collapse=""), "10")[[1]][,2]
}
df[as.vector(sapply(find_seq_str(),
function(n) {
i <- sum(dfr$lengths[1:(n-1)])
tail(i:(i+dfr$lengths[n]), -1)
})),]
# second approach
find_seq_ts <- function() {
which(apply(embed(dfr$values, 2), 1, function(x) all(x == c(0, 1))))
}
df[as.vector(sapply(find_seq_ts(),
function(n) {
i <- sum(dfr$lengths[1:(n)])+1
head(i:(i+dfr$lengths[n+1]), -1)
})),]
Both approaches need a run length encoding of the status vector.
The first does a single character replacement for -1 so we can make an unambiguous, contiguous string to then use str_locate to find the pairs that tell us when the target sequence starts then rebuilds the ranges of zeroes from the rle lengths.
If it needs to be base R I can try to whip up something with regexpr.
The second builds a paired matrix and compares for the same target sequence.
Caveats:
I did no benchmarking
Both create potentially big things if status is big.
I'm not completely positive it generalizes (hence my initial q).
David's is far more readable, maintainable & transferrable code but you get to deal with all the "goodness" that comes with using data.table ;-)
I wrapped the approaches in functions as they could potentially then be parameterized, but you could just as easily just assign the value to a variable or shove it into the sapply (ugh, tho).

Apply in R: recursive function that operates on its own previous result

How do I apply a function that can "see" the preceding result when operating by rows?
This comes up a lot, but my current problem requires a running total by student that resets if the total doesn't get to 5.
Example Data:
> df
row Student Absent Consecutive.Absences
1 A 0 0
2 A 1 1
3 A 1 2
4 A 0 0 <- resets to zero if under 5
5 A 0 0
6 A 1 1
7 A 1 2
8 A 1 3
9 B 1 1 <- starts over for new factor (Student)
10 B 1 2
11 B 0 0
12 B 1 1
13 B 1 2
14 B 1 3
15 B 1 4
16 B 0 0
17 B 1 1
18 B 1 2
19 B 1 3
20 B 1 4
21 B 1 5
22 B 0 5 <- gets locked at 5
23 B 0 5
24 B 1 6
25 B 1 7
I've tried doing this with a huge matrix of shifted vectors.
I've tried doing this with the apply family of functions and half of them do nothing, the other half hit 16GB of RAM and crash my computer.
I've tried straight looping and it takes 4+ hours (it's a big data set)
What bothers me is how easy this is in Excel. Usually R runs circles around Excel both in speed and writability, which leads me to believe I'm missing something elementary here.
Forgetting even the more challenging ("lock at 5") feature of this, I can't even get a cumsum that resets. There is no combination of factors I can think of to group for ave like this:
Consecutive.Absences = ave(Absent, ..., cumsum)
Obviously, grouping on Student will just give the Total Cumulative Absences -- it "remembers" the kid's absence over the gaps because of the split and recombine in ave.
So as I said, the core of what I don't know how to do in R is this:
How do I apply a function that can "see" the preceding result when operating by rows?
In Excel it would be easy:
C3 = IF($A3=$A2,$B3+$C2,$B3)*$B3
This excel function is displayed without the 5-absence lock for easy readability.
Once I figure out how to apply a function that looks at previous results of the same function in R, I'll be able to figure out the rest.
Thank you in advance for your help--this will be very useful in a lot of my applications!
Genuinely,
Sam
UPDATE:
Thank you everyone for the ideas on how to identify if a student has 5 consecutive absences!
However, that's easy enough to do in the database at the STUDENTS table. What I need to know is the number of consecutive absences by student in the attendance record itself for things like, "Do we count this particular attendance record when calculating other summary statistics?"
If you're looking to apply a function to every element in a vector while making use the previous element's value, you might want to check out "Reduce", with the accumulate parameter set to True
Here's an example:
##define your function that takes two parameters
##these are the 'previous' and the 'current' elements
runSum <- function(sum, x){
res = 0
if (x == 1){
res = sum + 1
}
else if (x == 0 & sum < 5){
res = 0
}
else{
res = sum
}
res
}
#lets look at the absent values from subject B
x = c(1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1)
Reduce(x=x, f=runSum, accumulate=T)
# [1] 1 2 0 1 2 3 4 0 1 2 3 4 5 5 5 6 7
It's fairly easy to identify the students with one or more runs of 5:
tapply(dfrm$Absent, dfrm$Student, function(x) rle(x)$value[rle(x)$length >=5] )
$A
integer(0)
$B
[1] 1
Look for any values of "1" in the result:
tapply(dfrm$Absent, dfrm$Student, function(x) 1 %in% rle(x)$value[rle(x)$length >=5] )
A B
FALSE TRUE
I also struggled through to a Reduce solution (but am second in priority to #kithpradhan):
ave(dfrm$Absent, dfrm$Student,
FUN= function(XX)
Reduce(function(x,y) if( x[1] >= 5){ y+x[1]
} else{ x[1]*y+y } , #Resets to 0 if y=0
XX, accumulate=TRUE)
)
#[1] 0 1 2 0 0 1 2 3 1 2 0 1 2 3 4 0 1 2 3 4 5 5 5 6 7
For the record, you can also create your own Reduce-derivative which receives f and x, and applies f(x) on its output until x == f(x) or maxiter is reached:
ireduce = function(f, x, maxiter = 50){
i = 1
while(!identical(f(x), x) & i <= maxiter) {x = f(x); i = i+1}; x
}

Resources