Residuals for Maximum-Likelihood and Quasi-Likelihood Models
Residuals of models fitted with functions betabin and negbin (formal class “glimML”), or
quasibin and quasipois (formal class “glimQL”).
## S4 method for signature 'glimML'
residuals(object, type = c("pearson", "response"), ...)
## S4 method for signature 'glimQL'
residuals(object, type = c("pearson", "response"), ...)object |
Fitted model of formal class “glimML” or “glimQL”. |
type |
Character string for the type of residual: “pearson” (default) or “response”. |
... |
Further arguments to be passed to the function, such as |
For models fitted with betabin or quasibin, Pearson's residuals are computed as:
(y - n * p.fit) / (n * p.fit * (1 - p.fit) * (1 + (n - 1) * φ))^{0.5}
where y and n are respectively the numerator and the denominator of the response, p.fit
is the fitted probability and φ is the fitted overdispersion parameter. When n = 0, the
residual is set to 0. Response residuals are computed as y/n - p.fit.
For models fitted with negbin or quasipois, Pearson's residuals are computed as:
(y - y.fit) / (y.fit + φ * y.fit^2)^{0.5}
where y and y.fit are the observed and fitted counts, respectively. Response residuals are computed as y - y.fit.
A numeric vector of residuals.
Matthieu Lesnoff matthieu.lesnoff@cirad.fr, Renaud Lancelot renaud.lancelot@cirad.fr
data(orob2)
fm <- betabin(cbind(y, n - y) ~ seed, ~ 1,
link = "logit", data = orob2)
#Pearson's chi-squared goodness-of-fit statistic
sum(residuals(fm, type = "pearson")^2)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.