glsdrm: Dose-response curve estimation by generalized least squares
In daniel-gerhard/medrc: Mixed effect dose-response curves

Description Usage Arguments Details Value Author(s) See Also Examples

Implements drc nonlinear functions into the nlme framework for nonlinear GLS dose-response modeling.

1 2	glsdrm(form, curveid=NULL, data, fct, correlation = NULL, weights = NULL, control = NULL, start = NULL)

`form`	Formula describing the dose-response relationship
`curveid`	Formula with parameter names on the left hand side (divided by +) and a column name in data, denoting a factor, to estimate separate parameters per factor-level. If NULL only fixed effects for a single curve will be estimated.
`data`	a data.frame object
`fct`	a function of the drc package
`correlation`	additional corClasses object
`weights`	additional varClasses object
`control`	list with nlme control arguments
`start`	optional list with initial values for the fixed components. If NULL the initial values will be found automatically.

An application of glsdrm is shown on the help pages of data broccoli. EDx and selectivity indices can be calculated with functions ED and EDcomp. Model-averaged ED can be computed by function mmaED.

An object of class glsdrc

Daniel Gerhard

drm, gls

library(nlme)
data(DNase)
DNase$Run <- factor(DNase$Run, levels=1:11)

ggplot(DNase, aes(y=density, x=conc, colour=Run)) + geom_point() + coord_trans(x="log")


#############
# fit a 5-parameter log-logistic model for each Run
mc <- glsdrm(density ~ conc, curveid=b + c + d + e + f ~ Run, 
             fct=LL.5(), data=DNase)
plot(mc, logx=TRUE, ndose=100)
print(mc)

# only curve specific inflection points
mcd <- glsdrm(density ~ conc, curveid=e ~ Run, 
              fct=LL.5(), data=DNase)
plot(mcd, logx=TRUE, ndose=100)
print(mcd)

# a 4-parameter log-Normal model
mcln <- glsdrm(density ~ conc, curveid=b + c + d + e ~ Run, 
               fct=LN.4(), data=DNase)
plot(mcln, logx=TRUE, ndose=100)

# AIC comparison
AIC(mc, mcd, mcln)


# ED25, ED50, ED75 estimation for LL.5 model
ED(mc, c(25, 50, 75), interval="tfls")

# pairwise comparison of ED50 between Runs
EDcomp(mc, c(50, 50))

# compound-symmetry correlation structure within Runs
mccs <- glsdrm(density ~ conc, fct=LL.5(), data=DNase, 
               correlation=corCompSymm(0.5, form=~1|Run))
summary(mccs)

# ED50 estimation
ED(mccs, c(25, 50, 75), interval="tfls")