Estimates the benchmark dose for averaged fixed effects of several medrc or glsdrc objects

mmaBMD(object, ..., respLev, ic = c("AIC", "BIC"), interval = c("none",
  "buckland", "kang"), level = 0.95, bmd = c("additional", "extra"),
  background = 0.05, marginal = FALSE, nGQ = 5, rfinterval = c(0, 1000))

Arguments

object

an medrc object

...

further medrc objects

respLev

a numeric vector containing the benchmark response levels

ic

the information criterion used for model averaging

interval

character string specifying the type of confidence intervals to be supplied.

level

confidence level, default at 0.95

bmd

benchmark dose estimation (smallest dose resulting in a probability of an abnormal response)

background

probability of an abnormal response

marginal

logical. If TRUE, marginalized BMD estimates are obtained. See BMDmarg for further information.

nGQ

integer. Specifies the number nof nodes for Gauss-Hermite quadrature.

rfinterval

numeric vector. Interval for root finding (uniroot) to search for ED values.