I am trying to reorder column names after expanding using the poly function in R.
The initial function I wrote looks like this:
> sorted_data
X0.1 X1.0 X0.2 X1.1 X2.0 X0.3 X1.2 X2.1 X3.0
1 1.701692 1.071616 2.895755 1.82356 1.148361 4.927683 3.103138 1.954156 1.230602
2 1.720578 1.035489 2.960388 1.78164 1.072238 5.093578 3.065450 1.844869 1.110291
s <- strsplit(substring(colnames(sorted_theta), 2), "\\.")
> s
[[1]]
[1] "0" "1"
[[2]]
[1] "1" "0"
[[3]]
[1] "0" "2"
[[4]]
[1] "1" "1"
[[5]]
[1] "2" "0"
colnames(sorted_data) <- sapply(s, function(x) {
vec <- c("x", "y", "z")[seq_along(x)]
x <- as.integer(x)
y <- rep(vec, rev(x))
paste(y, collapse = "")
})
colnames(sorted_data)
[1] "x" "y" "xx" "xy" "yy" "xxx" "xxy" "xyy" "yyy"
I am now trying to change the variable names to x1 x2, and x3. However, I was hoping to generalize the code to allow for more than just 3 variables. I also wanted to update it to use ^ for the powers such as the following:
sorted_data_test <- sorted_data
colnames(sorted_data_test) <- sapply(s, function(powers) {
terms <- mapply(function(power, index) {
if (power == "0") {
return(NULL)
} else if (power == "1") {
return(paste0("x", index))
} else {
return(paste0("x", index, "^", power))
}
}, powers, seq_along(powers), SIMPLIFY = FALSE)
# Filter out any NULL values from the terms list
terms <- Filter(Negate(is.null), terms)
# Collapse the terms into one string
paste(terms, collapse = "")
})
However, that gives:
print(colnames(sorted_theta_test))
[1] "x2" "x1" "x2^2" "x1x2" "x1^2" "x2^3" "x1x2^2" "x1^2x2" "x1^3"
How can I edit my second sapply to order the columns in the same way as the first sapply?
Thanks in advance.
I think you are almost there but just lack of rev outside seq_along(powers).
You can try
s <- strsplit(substring(colnames(sorted_data), 2), "\\.")
colnames(sorted_data_test) <- sapply(s, function(powers) {
terms <- mapply(function(power, index) {
if (power == "0") {
return(NULL)
} else if (power == "1") {
return(paste0("x", index))
} else {
return(paste0("x", index, "^", power))
}
}, powers, rev(seq_along(powers)), SIMPLIFY = FALSE) # <------ here is the minor change
# Filter out any NULL values from the terms list
terms <- Filter(Negate(is.null), terms)
# Sort terms alphabetically
sorted_terms <- sort(unlist(terms))
# Collapse the terms into one string
paste(sorted_terms, collapse = "")
})
and you will obtain
> sorted_data_test
x1 x2 x1^2 x2x1 x2^2 x1^3 x2x1^2 x2^2x1
1 1.701692 1.071616 2.895755 1.82356 1.148361 4.927683 3.103138 1.954156
2 1.720578 1.035489 2.960388 1.78164 1.072238 5.093578 3.065450 1.844869
x2^3
1 1.230602
2 1.110291