I'm still not entirely clear what you want, in particular what you
want to with factor variables ('squaring' them doesn't really seem to
make sense).
Some combination of sprintf/paste/reformulate might do what you want:
## get all second-order terms (doesn't square numeric values)
quad_terms
mula(paste("y ~", paste(result, collapse = " + ")))
model <- lda(formula, data = mydata)
Tim
-Original Message-
From: R-help On Behalf Of Stephen Bond via R-help
Sent: Monday, March 24, 2025 11:29 AM
To: Bert Gunter
Cc: r-help@r-project.org
Subject: Re: [R] how to create
@ Rui
that is the idea, but how do I apply this to a matrix with 200 columns?
I cannot write out the expression.
The colnames seem very messy, but they would be messy even under my
scheme with as little as 100 vars.
On Tue, 2025-03-25 at 06:15 +, Rui Barradas wrote:
> Às 15:28 de 24/03/2025,
Hello,
Sorry, much simpler:
poly(as.matrix(X), degree = 2L)
Hope this helps,
Rui Barradas
Às 13:58 de 25/03/2025, Rui Barradas escreveu:
Hello,
This seems to work and is independent of the number of columns. 'p' is
the output in my previous post.
f <- function(x, data = X) with(data,
Hello,
This seems to work and is independent of the number of columns. 'p' is
the output in my previous post.
f <- function(x, data = X) with(data, eval(parse(text = x)))
p2 <- poly(sapply(names(X), f), degree = 2L)
identical(p, p2)
# [1] TRUE
Hope this helps,
Rui Barradas
Às 13:42 de 2
Às 15:28 de 24/03/2025, Stephen Bond via R-help escreveu:
Folks,
I appreciate your effort, but maybe I was not explicit enough, so let
me try again.
The current setup for formulas does not allow for I(x^2) terms as
explained in the MASS book at the end of Section 6.2 the x:x
interaction is trea
Folks,
I appreciate your effort, but maybe I was not explicit enough, so let
me try again.
The current setup for formulas does not allow for I(x^2) terms as
explained in the MASS book at the end of Section 6.2 the x:x
interaction is treated as x.
So I need to write my own code, which is clumsy u
Full disclosure: I did not attempt to decipher your code.
But ~(A+B +C)^2 - (A + B + C)
gives all 2nd order interactions whether the terms are factors or numeric.
~I(A^2) + I(B^2) gives quadratics in A and B, which must be numeric, not
factors, of course
You can combine these as necessary to get
Depending on what your ultimate goal is, you could use something like
want <- model.matrix( ~ (a + b + c + d)^2 , data=your_data)
This will create a matrix with the appropriate main effects and first
order interactions. If you just want to run a simple regression, you
could do it directly
I am sending to this forum as stackoverflow has devolved into sth
pretty bad.
Below code shows how to get what I want in a clumsy way.
cols <- letters[1:4]
a1 <- outer(cols,cols,paste0)
b1 <- a1[!lower.tri(a1)]
X <- matrix(rnorm(80),ncol=4)
colnames(X) <- cols
X <- as.data.frame(X)
XX <- matrix(0
10 matches
Mail list logo
