| Type: | Package |
| Title: | Bootstrap Test for the Similarity of Dose Response Curves Concerning the Maximum Absolute Deviation |
| Version: | 1.1 |
| Author: | Kathrin Moellenhoff |
| Maintainer: | Kathrin Moellenhoff <kathrin.moellenhoff@rub.de> |
| Description: | Provides a bootstrap test which decides whether two dose response curves can be assumed as equal concerning their maximum absolute deviation. A plenty of choices for the model types are available, which can be found in the 'DoseFinding' package, which is used for the fitting of the models. See <doi:10.1080/01621459.2017.1281813> for details. |
| License: | GPL-3 |
| Depends: | lattice, DoseFinding, alabama |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 6.1.1 |
| NeedsCompilation: | no |
| Packaged: | 2019-09-11 13:16:39 UTC; Kathrin |
| Repository: | CRAN |
| Date/Publication: | 2019-09-11 13:50:02 UTC |
Implementation of Beta models
Description
Beta model:
m(d,\beta)=E_0+E_{max}B(\delta_1,\delta_2)(d/scal)^{\delta_1}(1-d/scal)^{\delta_2}
with
B(\delta_1,\delta_2)=(\delta_1+\delta_2)^{\delta_1+\delta_2}/(\delta_1^{\delta_1} \delta_2^{\delta_2})
and scal is a fixed dose scaling parameter.
Usage
betaMod(d, e, scal)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
scal |
fixed dose scaling parameter |
Value
Response value.
Bootstrap test for the equivalence of dose response curves via the maximum absolute deviation
Description
Function for testing whether two dose response curves can be assumed as equal concerning the hypotheses
H_0: \max_{d\in\mathcal{D}} |m_1(d,\beta_1)-m_2(d,\beta_2)|\geq \epsilon\ vs.\
H_1: \max_{d\in\mathcal{D}} |m_1(d,\beta_1)-m_2(d,\beta_2)|< \epsilon,
where
\mathcal{D}
denotes the dose range. See https://doi.org/10.1080/01621459.2017.1281813 for details.
Usage
bootstrap_test(data1, data2, m1, m2, epsilon, B = 2000, bnds1 = NULL,
bnds2 = NULL, plot = FALSE, scal = NULL, off = NULL)
Arguments
data1, data2 |
data frame for each of the two groups containing the variables referenced in dose and resp |
m1, m2 |
model types. Built-in models are "linlog", "linear", "quadratic", "emax", "exponential", "sigEmax", "betaMod" and "logistic" |
epsilon |
positive argument specifying the hypotheses of the test |
B |
number of bootstrap replications. If missing, default value of B is 5000 |
bnds1, bnds2 |
bounds for the non-linear model parameters. If not specified, they will be generated automatically |
plot |
if TRUE, a plot of the absolute difference curve of the two estimated models will be given |
scal, off |
fixed dose scaling/offset parameter for the Beta/ Linear in log model. If not specified, they are 1.2*max(dose) and 1 respectively |
Value
A list containing the p.value, the maximum absolute difference of the models, the estimated model parameters and the number of bootstrap replications. Furthermore plots of the two models are given.
References
https://doi.org/10.1080/01621459.2017.1281813
Examples
data(IBScovars)
male<-IBScovars[1:118,]
female<-IBScovars[119:369,]
bootstrap_test(male,female,"linear","emax",epsilon=0.35,B=300)
Implementation of absolute difference function
Description
Function calculating the absolute difference of two dose response models:
dff(d,\beta_1,\beta_2)=|m_1(d,\beta_1)-m_2(d,\beta_2)|
Usage
dff(d, beta1, beta2, m1, m2)
Arguments
d |
real-valued argument to the function (dose variable) |
beta1, beta2 |
model parameters (real vectors) |
m1, m2 |
model types. Built-in models are "linlog", "linear", "quadratic", "emax", "exponential", "sigEmax", "betaMod" and "logistic" |
Value
Response value for the absolute difference of two models.
Implementation of EMAX models
Description
Emax model:
m(d,\beta)=E_0+E_{max}\frac{d}{ED_{50}+d}
Usage
emax(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value.
Implementation of exponential models
Description
Exponential model:
m(d,\beta)=E_0+E_1(exp(d/\delta)-1)
Usage
exponential(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value.
Implementation of linear models
Description
Linear model:
m(d,\beta)=E_0+\delta d
Usage
linear(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value.
Implementation of linear in log models
Description
Linear in log Model model:
m(d,\beta)=E_0+\delta\ log(d+off)
and off is a fixed offset parameter.
Usage
linlog(d, e, off)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
off |
fixed offset parameter |
Value
Response value.
Implementation of logistic models
Description
Logistic model:
m(d,\beta)=E_0+\frac{E_{max}}{1+exp[(ED_{50}-d)/\delta]}
Usage
logistic(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value.
Implementation of quadratic models
Description
Quadratic model:
m(d,\beta)=E_0+\beta_1 d+\beta_2 d^2
Usage
quadratic(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value.
Implementation of Sigmoid Emax models
Description
Sigmoid Emax Model model:
m(d,\beta)=E_0+E_{max} \frac{d^h}{ED_{50}^h+d^h}
Usage
sigEmax(d, e)
Arguments
d |
real-valued argument to the function (dose variable) |
e |
model parameter |
Value
Response value