tests/testthat/test-Sacch2_1.R

In ArkaB-DS/nlsj: JN radical alternative to port of stats::nls()

# NLSProbName: Sacch2_1.R
# NLSProbDescription: {The Sacch2 data frame has 10 rows and 2 columns from an experiment on the pharmacokinetics of
# saccharin.
# The two columns are:This data frame contains the following columns:
# `time`: a numeric vector giving the time since drug administration (min).
# `conc`: a numeric vector giving the observed concentration of saccharin.

# }


# Use the Sacch2 data from NRAIA package

## DATA
time=c(0,   5,  15,  30,  45,  60,  75,  90, 105, 120)
conc = c(0.0, 184.3, 102.0,  50.5,  24.9,  14.1,   8.0,   5.7,   4.0,   2.9)
	
NLStestdata <- data.frame(time,conc)

## STARTING VALUE
Dose=1
lKa=13
lKe=17
lCl=7
NLSstart <- c(Dose=Dose,lKa=lKa,lKe=lKe,lCl=lCl) # a starting vector (named!)

## MODEL
NLSformula <-conc ~ Dose * exp(lKe+lKa-lCl) * (exp(-exp(lKe)*time) - exp(-exp(lKa)*time))/(exp(lKa) - exp(lKe))
NLSlower <- NULL
NLSupper <- NULL
NLSrunline <- "(formula=NLSformula, data=NLStestdata, start=NLSstart)"

# nls fails due to singular gradient
test_that("This is a singular gradient problem",
	expect_error(eval(parse(text=paste("nls",NLSrunline))),
		regex="singular gradient",ignore.case=TRUE)
	)
# nlsj fails due to singular jacobian
test_that("This is a singular gradient problem",
	expect_error(eval(parse(text=paste("nlsj::nlsj",NLSrunline))),
		regex="singular jacobian",ignore.case=TRUE)
	)