[R] Linear regression and stand deviation at the Linux command line

[Ivan Krylov]

В Thu, 22 Aug 2024 13:07:37 -0600
Keith Christian <keith1christian using gmail.com> пишет:

> I'm interested in R construct(s) to be entered at the command
> line that would output slope, y-intercept, and r-squared values read
> from a csv or other filename entered at the command line, and the same
> for standard deviation calculations, namely the standard deviation,
> variance, and z-scores for every data point in the file.

If you'd like to script R at the command line, consider the
commandArgs() function (try entering ?commandArgs at the R prompt).
This way you can pass a file path to an R process without unsafely
interpolating it into the R expression itself. These arguments can be
given to R --args or to Rscript (without the --args).

Also consider the 'littler' scripting-oriented R front-end
<https://CRAN.R-project.org/package=littler>, which puts the command
line arguments into the 'argv' variable and has a very convenient -d
option which loads CSV data from the standard input into a variable
named 'X'.

> Are line numbers, commas, etc. needed or no?

Depends on how you read it. By default, the function read.table() will
expect your data to be separated by a mixture of tabs and spaces and
will recognise a header if the first line contains one less column than
the rest of the file. Enter ?read.table at the R prompt to see the
available options (which include read.csv).

Good introductions to R include, well, "An Introduction to R" [1] (also
available by typing RShowDoc('R-intro') into the R prompt) and "Visual
Statistics" by Dr. A. Shipunov [2].

Start with functions read.table(), lm(), scale(), sd(), summary(). Use
str() to look at the structure of a variable: summary(lm(...)) will
return a named list from which you can extract the values you are
interested in (see ?summary.lm). When in doubt, call
help(name_of_the_function).

-- 
Best regards,
Ivan

[1]
https://cran.r-project.org/doc/manuals/R-intro.html

[2]
http://web.archive.org/web/20230106210646/http://ashipunov.info/shipunov/school/biol_240/en/visual_statistics.pdf