R help - Feb 2024 - Interactions in regression

Hi,
I do not want to make a plot, I try to make an output table in R, (in GUI
like Stata this is trivially easy task)
with regard to SO OP question. As I mentioned, in paper I would not do
this, but out of curiosity I use R this time trying to create it.
If in R this is trivial task as well, could you please show how to do it ?
After googling a lot I did not find solution.
regards,
Jacek

niedz., 25 lut 2024 o 17:00 Bert Gunter <bgunter.4567 at gmail.com>
napisa?(a):
> It is trivial in R to add whatever decorations to a plot that you would
> like, but that requires that you go beyond point and click production of
> graphics and write actual code. If you are unwilling or unable to do this,
> you are stuck with whatever various packaged graphics functionality
> provides.So you might want to search on "interaction plots for linear
> models in R" or similar at rseek.org or in your favorite web search
> engine if you haven't already done so. My minimal efforts brought up
lots
> of hits, though none may be useful for your concerns, especially, as has
> already been pointed out, as your query doesn't seem to make much sense
> statistically.
>
> Cheers,
> Bert
>
>
> On Sun, Feb 25, 2024 at 7:46?AM Jacek Kownacki <samsungx1298 at
gmail.com>
> wrote:
>
>> Hi All,
>> I stumbled upon some topics regarding interactions in anova and
regression
>> and packages for tabulating and visualizations the results of them.
>> Here we are:
>>
>>
https://stackoverflow.com/questions/77933272/how-to-add-a-reference-level-for-interaction-in-gtsummary-and-sjplot/77935742#77935742
>> ,
>>
>>
https://stackoverflow.com/questions/78016795/how-to-add-reference-levels-for-interaction-in-r?noredirect=1&lq=1
>> .
>> I was wondering because I usually use GUI software and these questions
did
>> not get answers, if from a technical point of view
>> how to do it, using these (sjPlot, gtsummary) or other ways to make
such
>> tables, inserting the reference levels of these mentioned interactions.
>> This is not likely to be used in publications (including three base
>> levels), but from the point of view of solving the topics this
questions
>> have interested me.
>> I tried myself to make it happen, but so far without success.
>> I recall this reprex based on SO:
>>
>> set.seed(1000)
>> my_data <- rbind(
>>   data.frame(time = "Pre", treatment = "Control",
response >> rnorm(100, mean=1)),
>>   data.frame(time = "Pre", treatment = "Treatment",
response >> rnorm(100, mean=2)),
>>   data.frame(time = "Post", treatment = "Control",
response >> rnorm(100, mean=1)),
>>   data.frame(time = "Post", treatment =
"Treatment", response >> rnorm(100, mean=2))
>> ) %>% mutate(time = factor(time, levels = c("Pre",
"Post")))
>> %>%mutate(treatment = factor(treatment, levels =
c("Control",
>> "Treatment")))
>> model3 <- lm(response ~ time * treatment, data = my_data)
>>
>> Thanks,
>> Jacek
>>
>>         [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
	[[alternative HTML version deleted]]

I have no real idea what you are trying to do, but if a table is
what you want, you can probably get it using the table() function.
Or, more likely, the xtabs() function.

Using your example from an earlier post (adjusted to make it
comprehensible to the human mind):

set.seed(1000)
time      <- factor(rep(c("Pre","Post"),each=200))
treatment <-
factor(rep(rep(c("Control","Treatment"),each=100),2))
mu        <- rep(rep(1:2,each=100),2)
response  <- rnorm(400,mean=mu)
xmpldata  <- data.frame(time=time,treatment=treatment,response)
mod       <- lm(response~time*treatment,data=xmpldata)
yhat      <- fitted(mod)
xtb       <- with(xmpldata,xtabs(yhat ~ time+treatment))

print(xtb)
>       treatment
> time     Control Treatment
>   Post  94.10501 201.99112
>   Pre  101.63792 210.04248
Is that (something like) what you want?

cheers,

Rolf Turner

P.S. You said:  "I usually use GUI software".  Now *there* lies your
problem.  A GUI is a black box that removes all control over what is
going on, from your hands.

R. T.




-- 
Honorary Research Fellow
Department of Statistics
University of Auckland
Stats. Dep't. (secretaries) phone:
         +64-9-373-7599 ext. 89622
Home phone: +64-9-480-4619