Transform a numeric variable using a GLM link function, or return an estimate of same.
glt(x, family = NULL, force.est = FALSE, ...)
a positive numeric vector, corresponding to a variable to be transformed. Can also be a list or nested list of such vectors.
Optional, the error distribution family containing the link
function which will be used to transform x
(see
family
for specification details). If not supplied, it
is determined from x
(see Details).
Logical, whether to force the return of the estimated rather than direct transformation, where the latter can be produced (i.e. does not contain undefined values).
Not currently used.
A numeric vector of the transformed values, or an array, list of vectors/arrays, or nested list.
glt
can be used to provide a 'generalised' transformation of
a numeric variable using the link function from a generalised linear model
(GLM). The transformation is generalised in the sense that it can always be
produced, even where a standard link transformation would produce undefined
values. It achieves this via an estimate based on the 'working' response
variable of the GLM (see below). If the error distribution family
is
not specified (default), then it is determined (roughly) from x
,
with binomial(link = "logit")
used when all x <= 1 and
poisson(link = "log")
otherwise. Although the function is generally
intended for binomial or poisson variables, any variable which can be fit
using glm
can be supplied. One of the key purposes of glt
is
to allow the calculation of fully standardised model coefficients for GLMs
(in which case x
= the response variable), while it can also
facilitate the proper calculation of SEM indirect effects (see below).
Estimating the link transformation
A key challenge in generating fully standardised model coefficients for a GLM with a non-gaussian link function is the difficulty in calculating appropriate standardised ranges (typically the standard deviation) for the response variable in the link scale. This is because directly transforming the response will often produce undefined values. Although methods for circumventing this issue by indirectly estimating the variance of the link-transformed response have been proposed - including a latent-theoretic approach for binomial models (McKelvey & Zavoina 1975) and a more general variance-based method using a pseudo R-squared (Menard 2011) - here an alternative approach is used. Where transformed values are undefined, the function will instead return the synthetic 'working' response from the iteratively reweighted least squares algorithm (IRLS) of the GLM (McCullagh & Nelder 1989). This is reconstructed as the sum of the linear predictor and the working residuals - with the latter comprising the error of the model in the link scale. The advantage of this approach is that a relatively straightforward 'transformation' of any non-gaussian response is readily attainable in all cases. The standard deviation (or other relevant range) can then be calculated using values of the transformed response and used to scale the coefficients. An additional benefit for piecewise SEMs is that the transformed rather than original response can then be specified as a predictor in other models, ensuring that standardised indirect and total effects are calculated correctly (i.e. using the same units).
To ensure a high level of 'accuracy' in the working response - in the sense that the inverse-transformation is practically indistinguishable from the original response variable - the function uses the following iterative fitting procedure to calculate a 'final' working response:
The working response is extracted from this model.
The inverse transformation of the working response is then calculated.
If the inverse transformation is 'effectively' equal to
the original response (tested using all.equal
), the working response
is returned; otherwise, the GLM is re-fit with the working response now as
the predictor, and steps 2-4 are repeated - each time with an additional
IWLS iteration.
This approach will generate a very reasonable transformation of the response variable, which will also be practically indistinguishable from the direct transformation (where this can be compared - see Examples). It also ensures that the transformed values, and hence the standard deviation, are the same for any GLM fitting the same response - provided it uses the same link function - facilitating model comparisons, selection, and averaging.
Grace, J.B., Johnson, D.J., Lefcheck, J.S. and Byrnes, J.E.K. (2018) Quantifying relative importance: computing standardized effects in models with binary outcomes. Ecosphere 9, e02283. https://doi.org/gdm5bj
McCullagh P. and Nelder, J. A. (1989) Generalized Linear Models (2nd Edition). London: Chapman and Hall.
McKelvey, R. D., & Zavoina, W. (1975). A statistical model for the analysis of ordinal level dependent variables. The Journal of Mathematical Sociology, 4(1), 103-120. https://doi.org/dqfhpp
Menard, S. (2011) Standards for Standardized Logistic Regression Coefficients. Social Forces 89, 1409-1428. https://doi.org/bvxb6s
# NOT RUN {
## Compare estimate with a direct link transformation
## (test with a poisson variable, log link)
set.seed(1)
y <- rpois(30, lambda = 10)
yl <- glt(y, force.est = TRUE)
## Effectively equal?
all.equal(log(y), yl, check.names = FALSE)
# TRUE
## Actual difference...
all.equal(log(y), yl, check.names = FALSE, tolerance = .Machine$double.eps)
# "Mean relative difference: 1.05954e-12"
# }
Run the code above in your browser using DataLab