[R] Using functions within functions (environment problems)

From: Colin Beale <c.beale_at_macaulay.ac.uk>
Date: Fri 26 Jan 2007 - 09:58:33 GMT

Hi everyone,

I've been having difficulty writing wrapper functions for some functions where those same functions include other functions with eval()
calls where the environment is specified. A very simple example using function lmer from lme4:

lmerWrapper <- function(formula, data, family = gaussian, method = c("REML",

    "ML", "PQL", "Laplace", "AGQ"), control = list(), start = NULL,     subset, weights, na.action, offset, contrasts = NULL, model = TRUE,     ...)

    xNew <- runif(0,1, length(data[,1]))
    fNew <- sample(1:4, length(data[,1]), replace = T)
    data <- as.data.frame(cbind(data, xNew, fNew))
    formula <- update(formula, .~. + xNew + (1|fNew))     out <- lmer (formula = formula, data = data, family = family, method =
        method, weights = weights, control = control, start = start,
        subset = subset, na.action = na.action, offset = offset,
        contrasts = contrasts, model = model)

dat <- data.frame(Y = rnorm(100), X1 = rnorm(100), X2 = rnorm(100),

        F1 = as.factor(sample(1:4, 400, replace = T))) test <- lmerWrapper (Y ~ X1 + X2 + (1|F1), data = dat)

This function attempts to create two new variables, add these to the data.frame and modify the formula for include these two new variables, and then fit this expanded model using lmer. Clearly the example is silly, but it illustrates the problem as it fails with the error:

Error in eval(expr, envir, enclos) : object "xNew" not found

because a function within lmer (lmerFrames) makes an eval call where the environment is specified as the .GlobalEnv which doesn't contain xNew - it needs to be looking in the environment from which it was called rather than going right back to the root. In a more general context, I might like to create a function where I don't specify a new data.frame in the wrapper, but want the function to search the back down
the search path for each component, finding xNew in the environment of the wrapper function and X1 in the .GlobalEnv. Is there a general solution to this? Or do I need to create a modified lmer function that calls a modified lmerFrames function where I specify the environment differently? There's quite a lot in the archives dealing with environments and search paths, but I'm either not understanding how to apply the wisdom therein, or not finding the specific answer to my problem, and I should say that I have this problem with more than just the lmer function.

Thanks for any pointers,


> sessionInfo()

R version 2.4.1 (2006-12-18)

LC_COLLATE=English_United Kingdom.1252;LC_CTYPE=English_United Kingdom.1252;LC_MONETARY=English_United
Kingdom.1252;LC_NUMERIC=C;LC_TIME=English_United Kingdom.1252

attached base packages:
[1] "stats" "graphics" "grDevices" "datasets" "tcltk" "utils"

    "methods" "base"

other attached packages:

      debug mvbutils lme4 Matrix lattice svSocket

       svIO      R2HTML      svMisc       svIDE 
    "1.1.0"     "1.1.1" "0.9975-10"  "0.9975-8"   "0.14-16"    
    "0.9-5"      "1.58"     "0.9-5"     "0.9-5" 

Dr. Colin Beale
Spatial Ecologist
The Macaulay Institute

AB15 8QH

Tel: 01224 498245 ext. 2427
Fax: 01224 311556
Email: c.beale@macaulay.ac.uk

Please note that the views expressed in this e-mail are those of the
sender and do not necessarily represent the views of the Macaulay
Institute. This email and any attachments are confidential and are
intended solely for the use of the recipient(s) to whom they are
addressed. If you are not the intended recipient, you should not read,
copy, disclose or rely on any information contained in this e-mail, and
we would ask you to contact the sender immediately and delete the email
from your system. Thank you.
Macaulay Institute and Associated Companies, Macaulay Drive,
Craigiebuckler, Aberdeen, AB15 8QH.

R-help@stat.math.ethz.ch mailing list
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Fri Jan 26 21:05:07 2007

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Fri 26 Jan 2007 - 13:30:30 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.