metadrm: 2-stage estimation in hierarchical dose-response models via...
In DoseResponse/medrc: Mixed effect dose-response curves

Description Usage Arguments Details Value

A two-stage meta-analysis approach to fit hierarchical dose-response models.

1 2	metadrm(formula, fct, ind, data, type = "continuous", cid2 = NULL, pms2 = NULL, struct = "UN", method = "REML", ...)

`formula`	a symbolic description of the model to be fit. Either of the form 'response ~ dose' or as a data frame with response values in first column and dose values in second column.
`fct`	a list with three or more elements specifying the non-linear function, the accompanying self starter function, the names of the parameter in the non-linear function and, optionally, the first and second derivatives as well as information used for calculation of ED values. Currently available functions include, among others, the four- and five-parameter log-logistic models `LL.4`, `LL.5` and the Weibull model `W1.4`. Use `getMeanFunctions` for a full list.
`ind`	a factor identifying individual curves.
`data`	an optional data frame containing the variables in the model.
`type`	a character string specifying the distribution of the data (parameter estimation will depend on the assumed distribution as different log likelihood functions will be used). The default is "continuous", corresponding to assuming a normal distribution. The choices "binary" and "event" imply a binomial and multinomial distribution, respectively. The choice "ssd" is for fitting a species sensitivity distribution.
`cid2`	a numeric vector or factor containing the between-curve grouping of the data.
`pms2`	a list containing a formula for each parameter in the nonlinear function.
`struct`	correlation structure between population coefficients, either "CS" for compound symmetry, "HCS" for heteroscedastic compound symmetry, "UN" for an unstructured variance-covariance matrix, "ID" for a scaled identity matrix, "DIAG" for a diagonal matrix, "AR" for an AR(1) autoregressive structure, or "HAR" for a heteroscedastic AR(1) autoregressive structure.
`method`	character string specifying whether the meta-analysis model should be fitted via maximum-likelihood ("ML") or via restricted maximum-likelihood ("REML") estimation. Default is "REML".
`...`	further arguments supplied to function drm.