[R] How to get variable name while doing series of regressions in an automated manner?

[Marc Schwartz]

> On Oct 27, 2015, at 10:19 AM, Ravi Varadhan <ravi.varadhan at jhu.edu> wrote:
> 
> Hi,
> 
> I am running through a series of regression in a loop as follows:
> 
> results <- vector("list", length(mydata$varnames))
> 
> for (i in 1:length(mydata$varnames)) {
> results[[i]] <- summary(lm(log(eval(parse(text=varnames[i]))) ~ age + sex + CMV.status, data=mydata))
> }
> 
> Now, when I look at the results[i]] objects, I won't be able to see the original variable names.  Obviously, I will only see the following:
> 
> Call:
> lm(formula = log(eval(parse(text = varnames[i]))) ~ age + sex + CMV.status,
>    data = mydata)
> 
> 
> Is there a way to display the original variable names on the LHS?  In addition, is there a better paradigm for doing these type of series of regressions in an automatic fashion?
> 
> Thank you very much,
> Ravi


Ravi,

Something like this, using the 'iris' dataset might be helpful as an example:

# Define the response variables
VarNames <- c("Sepal.Length", "Sepal.Width", "Petal.Length")

# Create the formulae
> paste0("log(", VarNames, ") ~ Petal.Width + Species")
[1] "log(Sepal.Length) ~ Petal.Width + Species"
[2] "log(Sepal.Width) ~ Petal.Width + Species" 
[3] "log(Petal.Length) ~ Petal.Width + Species"


# Create a list of model summary objects
# The result of paste0() will be coerced to a formula by lm()
# if a valid formula, so no need to call as.formula()
MODS <- lapply(paste0("log(", VarNames, ") ~ Petal.Width + Species"), 
               function(x) summary(lm(x, data = iris)))


You can either use the original 'VarNames' vector for the source response variables, or consider:

> as.character(formula(MODS[[1]]))
[1] "~"                     "log(Sepal.Length)"    
[3] "Petal.Width + Species"


> sapply(MODS, function(x) formula(x)[[2]])
[[1]]
log(Sepal.Length)

[[2]]
log(Sepal.Width)

[[3]]
log(Petal.Length)


Regards,

Marc Schwartz