randomize or permuting values in a data.frame - r

I have a data.frame that looks like this: (my real dataframe is bigger):
df <- data.frame(A=c("a","b","c","d","e","f","g","h","i"),
B=c("1","1","1","2","2","2","3","3","3"),
C=c(0.1,0.2,0.4,0.1,0.5,0.7,0.1,0.2,0.5))
> df
A B C
1 a 1 0.1
2 b 1 0.2
3 c 1 0.4
4 d 2 0.1
5 e 2 0.5
6 f 2 0.7
7 g 3 0.1
8 h 3 0.2
9 i 3 0.5
I want to add several n-columns (something similar to permutations) where the column D would be a random value from df$C but this value should only be picked from those rows with the dame value of df$B, an example of the desired output would be:
df <- data.frame(A=c("a","b","c","d","e","f","g","h","i"),
B=c("1","1","1","2","2","2","3","3","3"),
C=c(0.1,0.2,0.4,0.1,0.5,0.7,0.1,0.2,0.5),
D=c(0.2,0.2,0.1,0.5,0.7,0.1,0.5,0.5,0.2))
> df
A B C D
1 a 1 0.1 0.2
2 b 1 0.2 0.2
3 c 1 0.4 0.1
4 d 2 0.1 0.5
5 e 2 0.5 0.7
6 f 2 0.7 0.1
7 g 3 0.1 0.5
8 h 3 0.2 0.5
9 i 3 0.5 0.2
I've tried with plyr package but my approach does not work properly:
ddply(df, levels(.(B)), transform, D=sample(C))
I also have thought about splitting the dataframe based on df$B and then using a function to add the column in each dataframe using lapply however I have no clue how select for the levels of df$B,
Many thanks

No need for plyr, ave will do the trick.
transform(df, D=ave(C, B, FUN=function(b) sample(b, replace=TRUE)))

Related

Removing outliers above 3SD above the mean of a time series profile

Removing outliers (by column) above 3 standard deviations of the median in R with multiple columns in a time series. I want to remove the row that has an outlier.
In the example below, the last row would be removed because there is an outlier in column B.
See example data and output
Example data
A B C
1 0.1 2
2 0.2 3
3 0.3 4
4 0.4 5
5 8.0 6
Example output
A B C
1 0.1 2
2 0.2 3
3 0.3 4
4 0.4 5
You should probably base your cut on the median rather than the mean!
> d <- data.frame(A=1:5, B=c(0.1*(1:4),8), C=2:6)
> cut <- apply(d, 2, quantile, 0.997)
> sel <- apply(d, 1, function(x) all(x<cut))
> d[sel,]
A B C
1 1 0.1 2
2 2 0.2 3
3 3 0.3 4
4 4 0.4 5

Use conditions from multiple variables to replace a variable in R

I did some searches but could not find the best keywords to phrase my question so I think I will attempt to ask it here.
I am dealing with a data frame in R that have two variables represent the identity of the data points. In the following example, A and 1 represent the same individual, B and 2 are the same and so are C and 3 but they are being mixed in the original data.
ID1 ID2 Value
A 1 0.5
B 2 0.8
C C 0.7
A A 0.6
B 2 0.3
3 C 0.4
2 2 0.3
1 A 0.4
3 3 0.6
What I want to achieve is to unify the identity by using only one of the identifiers so it can be either:
ID1 ID2 Value ID
A 1 0.5 A
B 2 0.8 B
C C 0.7 C
A A 0.6 A
B 2 0.3 B
3 C 0.4 C
2 2 0.3 B
1 A 0.4 A
3 3 0.6 C
or:
ID1 ID2 Value ID
A 1 0.5 1
B 2 0.8 2
C C 0.7 3
A A 0.6 1
B 2 0.3 2
3 C 0.4 3
2 2 0.3 2
1 A 0.4 1
3 3 0.6 3
I can probably achieve it by using ifelse function but that means I have to write two ifelse statements for each condition and it does not seem efficient so I was wondering if there is a better way to do it. Here is the example data set.
df=data.frame(ID1=c("A","B","C","A","B","3","2","1","3"),
ID2=c("1","2","C","A","2","C","2","A","3"),
Value=c(0.5,0.8,0.7,0.6,0.3,0.4,0.3,0.4,0.6))
Thank you so much for the help!
Edit:
To clarify, the two identifiers I have in my real data are longer string of texts instead of just ABC and 123. Sorry I did not make it clear.
An option is to to detect the elements that are only digits, convert to integer, then get the corresponding LETTERS in case_when
library(dplyr)
library(stringr)
df %>%
mutate(ID = case_when(str_detect(ID1, '\\d+')~
LETTERS[as.integer(ID1)], TRUE ~ ID1))
# ID1 ID2 Value ID
#1 A 1 0.5 A
#2 B 2 0.8 B
#3 C C 0.7 C
#4 A A 0.6 A
#5 B 2 0.3 B
#6 3 C 0.4 C
#7 2 2 0.3 B
#8 1 A 0.4 A
#9 3 3 0.6 C
Or more compactly
df %>%
mutate(ID = coalesce(LETTERS[as.integer(ID1)], ID1))
If we have different sets of values, then create a key/value dataset and do a join
keyval <- data.frame(ID1 = c('1', '2', '3'), ID = c('A', 'B', 'C'))
left_join(df, keyval) %>% mutate(ID = coalesce(ID, ID1))
A base R option using replace
within(
df,
ID <- replace(
ID1,
!ID1 %in% LETTERS,
LETTERS[as.numeric(ID1[!ID1 %in% LETTERS])]
)
)
or ifelse
within(
df,
ID <- suppressWarnings(ifelse(ID1 %in% LETTERS,
ID1,
LETTERS[as.integer(ID1)]
))
)
which gives
ID1 ID2 Value ID
1 A 1 0.5 A
2 B 2 0.8 B
3 C C 0.7 C
4 A A 0.6 A
5 B 2 0.3 B
6 3 C 0.4 C
7 2 2 0.3 B
8 1 A 0.4 A
9 3 3 0.6 C

subsetting large data frames with condition

I have got the following dataset:
ID s1 s2 s3
A 0.6 1 0.3
B 3 0.4 0.4
C 3 2 1
D 0 0.3 0.2
E 3 2 0.1
i would like to retain the rows which have the value >=0.5 at least two of the 3 samples
So, the new data frame would be:
ID s1 s2 s3
A 0.6 1 0.3
C 3 2 1
E 3 2 0.1
Thanks in advance
You can do
df[rowSums(df[-1] > 0.5) >= 2, ]
# ID s1 s2 s3
#1 A 0.6 1 0.3
#3 C 3.0 2 1.0
#5 E 3.0 2 0.1
We create a logical matrix df[-1] > 0.5 and check if at least two values per row are TRUE.
data
df <- read.table(text="ID s1 s2 s3
A 0.6 1 0.3
B 3 0.4 0.4
C 3 2 1
D 0 0.3 0.2
E 3 2 0.1", header = TRUE, stringsAsFactor = FALSE)

How to reset row names?

Here is a sample data set:
sample1 <- data.frame(Names=letters[1:10], Values=sample(seq(0.1,1,0.1)))
When I'm reordering the data set, I'm losing the row names order
sample1[order(sample1$Values), ]
Names Values
7 g 0.1
4 d 0.2
3 c 0.3
9 i 0.4
10 j 0.5
5 e 0.6
8 h 0.7
6 f 0.8
1 a 0.9
2 b 1.0
Desired output:
Names Values
1 g 0.1
2 d 0.2
3 c 0.3
4 i 0.4
5 j 0.5
6 e 0.6
7 h 0.7
8 f 0.8
9 a 0.9
10 b 1.0
Try
rownames(Ordersample2) <- 1:10
or more generally
rownames(Ordersample2) <- NULL
I had a dplyr usecase:
df %>% as.data.frame(row.names = 1:nrow(.))

Fill nth columns in a dataframe

I have this data frame:
df <- data.frame(A=c("a","b","c","d","e","f","g","h","i"),
B=c("1","1","1","2","2","2","3","3","3"),
C=c(0.1,0.2,0.4,0.1,0.5,0.7,0.1,0.2,0.5))
> df
A B C
1 a 1 0.1
2 b 1 0.2
3 c 1 0.4
4 d 2 0.1
5 e 2 0.5
6 f 2 0.7
7 g 3 0.1
8 h 3 0.2
9 i 3 0.5
I would like to add 1000 further columns and fill this columns with the values generated by :
transform(df, D=ave(C, B, FUN=function(b) sample(b, replace=TRUE)))
I've tried with a for loop but it does not work:
for (i in 4:1000){
df[, 4:1000] <- NA
df[,i] = transform(df, D=ave(C, B, FUN=function(b) sample(b, replace=TRUE)))
}
For efficiency reasons, I suggest running sample only once for each group. This can be achieved with this:
sample2 <- function(x, size)
{
if(length(x)==1) rep(x, size) else sample(x, size, replace=TRUE)
}
new_df <- do.call(rbind, by(df, df$B,
function(d) cbind(d, matrix(sample2(d$C, length(d$C)*1000),
ncol=1000))))
Notes:
I've created sample2 in case there is a group with only one C value. Check ?sample to see what I mean.
The names of the columns will be numbers, from 1 to 1000. This can be changed as in the answer by #agstudy.
The row names are also changed. "Fixing" them is similar, just use row.names instead of col.names.
Using replicate for example:
cbind(df,replicate(1000,ave(df$C, df$B,
FUN=function(b) sample(b, replace=TRUE))))
To add 4 columns for example:
cbind(df,replicate(4,ave(df$C, df$B,
FUN=function(b) sample(b, replace=TRUE))))
A B C 1 2 3 4
1 a 1 0.1 0.2 0.2 0.1 0.2
2 b 1 0.2 0.4 0.2 0.4 0.4
3 c 1 0.4 0.1 0.1 0.1 0.1
4 d 2 0.1 0.1 0.5 0.5 0.1
5 e 2 0.5 0.7 0.1 0.5 0.1
6 f 2 0.7 0.1 0.7 0.7 0.7
7 g 3 0.1 0.2 0.5 0.2 0.2
8 h 3 0.2 0.2 0.1 0.2 0.1
9 i 3 0.5 0.5 0.5 0.1 0.5
Maybe you need to rename columns by something like :
gsub('([0-9]+)','D\\1',colnames(res))
1] "A" "B" "C" "D1" "D2" "D3" "D4"

Resources