A simple diagnostic of whether the maximum likelihood estimates are infinite
A simple diagnostic of whether the maximum likelihood estimates are infinite
## S3 method for class 'glm' check_infinite_estimates(object, nsteps = 20, ...)
object |
the result of a |
nsteps |
starting from |
... |
currently not used. |
check_infinite_estimates attempts to identify the occurrence
of infinite estimates in GLMs with binomial responses by
successively refitting the model. At each iteration the maximum
number of allowed IWLS iterations is fixed starting from 1 to
nsteps (by setting control = glm.control(maxit = j),
where j takes values 1, ..., nsteps in
glm). For each value of maxit, the estimated
asymptotic standard errors are divided to the corresponding ones
from control = glm.control(maxit = 1). Then, based on the
results in Lesaffre & Albert (1989), if the sequence of ratios in
any column of the resultant matrix diverges, then complete or
quasi-complete separation occurs and the maximum likelihood
estimate for the corresponding parameter has value minus or plus
infinity.
check_infinite_estimates will be removed from brglm2
at version 0.8. An new version of check_infinite_estimates
is now maintained in the detectseparation R package at
https://cran.r-project.org/package=detectseparation. In order
to use the version in detect_separation load first
brglm2 and then detectseparation, i.e.
library(brglm2); library(detectseparation).
Kosmidis I, Firth D (2020). Jeffreys-prior penalty, finiteness and shrinkage in binomial-response generalized linear models. *Biometrika* doi: 10.1093/biomet/asaa052
Lesaffre E, Albert A (1989). Partial Separation in Logistic Discrimination. *Journal of the Royal Statistical Society. Series B (Methodological)*, **51**, 109-116 doi: 10.1111/j.2517-6161.1989.tb01752.x
## endometrial data from Heinze \& Schemper (2002) (see ?endometrial)
data("endometrial", package = "brglm2")
endometrialML <- glm(HG ~ NV + PI + EH, data = endometrial,
family = binomial("probit"))
## clearly the maximum likelihood estimate for the coefficient of
## NV is infinite
check_infinite_estimates(endometrialML)
## Not run:
## Aligator data (Agresti, 2002, Table~7.1)
data("alligator", package = "brglm2")
all_ml <- brmultinom(foodchoice ~ size + lake , weights = round(freq/3),
data = alligators, type = "ML", ref = 1)
## Clearly some estimated standard errors diverge as the number of
## Fisher scoring iterations increases
matplot(check_infinite_estimates(all_ml), type = "l", lty = 1,
ylim = c(0.5, 1.5))
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.