access the formula interface

lm_with_flexible_variables

``` r
# with a formula
fit_something <- function(data, formula, ...) {
  lm(as.formula(formula), data = data, ...)
}

fit_something(mtcars, "cyl ~mpg")
#> 
#> Call:
#> lm(formula = as.formula(formula), data = data)
#> 
#> Coefficients:
#> (Intercept)          mpg  
#>     11.2607      -0.2525


# with separate inputs for response and predictors
fit_something2 <- function(data, response, predictors, ...) {
  formula <- as.formula(paste(response, " ~ ", paste(predictors, collapse = "+")))
  lm(formula, data = data, ...)
}

fit_something2(mtcars, "cyl", setdiff(names(mtcars), "cyl"))
#> 
#> Call:
#> lm(formula = formula, data = data)
#> 
#> Coefficients:
#> (Intercept)          mpg         disp           hp         drat  
#>   12.107199    -0.004857     0.004610     0.003723    -0.427435  
#>          wt         qsec           vs           am         gear  
#>   -0.222489    -0.187945    -0.644076    -0.500770    -0.500323  
#>        carb  
#>    0.179872


# further arguments passed to lm via ...  see ?lm
fit_something2(data = mtcars, response = "cyl", 
               predictors = setdiff(names(mtcars), "cyl"), 
               weights = abs(rnorm(nrow(mtcars))))
#> 
#> Call:
#> lm(formula = formula, data = data, weights = ..1)
#> 
#> Coefficients:
#> (Intercept)          mpg         disp           hp         drat  
#>   13.812036    -0.021711    -0.001586     0.003722    -0.896883  
#>          wt         qsec           vs           am         gear  
#>    0.549706    -0.370025    -1.096041    -1.079088     0.643830  
#>        carb  
#>   -0.215578
```