[R] [External] Creating model formulas programmatically

[Duncan Murdoch]

On 2025-03-29 10:30 p.m., Bert Gunter wrote:
> As always, I would like to thank all who responded for their insights and
> suggestions. I have learned from them.
> 
> Thus far, my own aesthetic preference -- and therefore not to be considered
> in any sense as a "best" approach -- is to use Duncan's suggestion to
> produce the call directly with call() rather than substitute in my simple
> for() loop; i.e.
> 
> somenames <- c("Heigh", "Ho", "Silver", "Away")
> nms <- lapply(somenames, as.name)
> frm<- nms[[1]]
> for(x in nms[-1]) frm <<- call("+", frm, x)
> frm <-bquote(~ (.(frm))^2, list(frm =frm))
> 
> ## which yields
> 
>> frm
> ~(Heigh + Ho + Silver + Away)^2

A couple of comments:

1. Using <<- in the for loop is a bad idea.  It works fine at the top 
level, but if you put that code in a function, it would try to make the 
assignment outside the scope of the function.  Standard <- assignment is 
better here.

2. You can use Reduce() to replace the for loop.  Some people would 
prefer that:

   Reduce(\(x,y) call("+", x, y), nms)

produces the same result as the loop.

Duncan Murdoch

> 
> I obviously still did not make make my requirements clear, since after
> converting the character string names into language onjects via as.name(),
> the above (and Duncan's response and my original query) do not make any use
> of string parsing/conversion. That is what I meant. As respondents pointed
> out, there are some base R functions that put the string manipulations
> behind a curtain so that one need not directly worry about them, but I
> didn't want to go there either. Again, my apologies for my imprecision.
> 
> Finally, I still welcome any other approaches along these lines and
> especially would appreciate anyone pointing out any fatal flaws or even
> modest infelicities in the above approach.
> 
> Cheers,
> Bert
> 
> "An educated person is one who can entertain new ideas, entertain others,
> and entertain herself."
> 
> 
> 
> On Sat, Mar 29, 2025 at 6:32 PM Ebert,Timothy Aaron <tebert using ufl.edu> wrote:
> 
>> I am confused. Richard's answer that Bert did not like did not use parse
>> explicitly. Richard pasted together a string that a function like lm() will
>> have to parse to run the analysis. However, the answers so far do not use
>> parse(). In the reply to Richard, Bert indicated we cannot use strings.
>> Even if I pass a vector where R can assume that the first variable is the
>> dependent variable and all others are independent variables, I see no way
>> of processing that within the current modeling system without generating a
>> formula (a string) that gets parsed.
>>
>> -----Original Message-----
>> From: R-help <r-help-bounces using r-project.org> On Behalf Of Richard M.
>> Heiberger
>> Sent: Saturday, March 29, 2025 8:47 PM
>> To: Bert Gunter <bgunter.4567 using gmail.com>
>> Cc: R-help <R-help using r-project.org>
>> Subject: Re: [R] [External] Creating model formulas programmatically
>>
>> [External Email]
>>
>> my take of the assignment was to avoid 'parse' specifically.
>> we start with a character vector, so avoiding characters is not possible.
>> i was dealing with the fortune "if parse is the answer, you have the wrong
>> question"
>>
>> Sent from my iPhone
>>
>> On Mar 29, 2025, at 15:39, Bert Gunter <bgunter.4567 using gmail.com> wrote:
>>
>> 
>> Thanks, Rich.
>>
>> I thought of that, too, but it violates the spirit of my restraints (to
>> avoid character strings), which I unfortunately did not clearly articulate.
>> So my apologies for that failure. My concern is that with more complex
>> model formula, using as.formula, etc. to parse/convert character strings
>> can get a bit hairy. But in most cases, as here maybe, it may be perfectly
>> fine. So think of my post as mostly my attempt to learn some new tricks
>> rather than to solve a useful problem. I hope this is not unfair to the
>> list.
>>
>> Cheers,
>> Bert
>>
>>
>>
>>
>>
>> On Sat, Mar 29, 2025 at 3:1 PM Richard M. Heiberger <rmh using temple.edu
>> <mailto:rmh using temple.edu>> wrote:
>>> somenames <- c("Heigh", "Ho", "Silver", "Away")
>>> as.formula(paste("~(",paste(somenames, collapse="+"),")^2"))
>> ~(Heigh + Ho + Silver + Away)^2
>>>
>>
>>> On Mar 29, 2025, at 14:30, Bert Gunter <bgunter.4567 using gmail.com<mailto:
>> bgunter.4567 using gmail.com>> wrote:
>>>
>>> somenames <- c("Heigh", "Ho", "Silver", "Away")
>>
>>
>>
>>          [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> https://www.r-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.