- Fitted values and residuals have now the correct length even when there are zero weights.
- Fixed a bug when calculating the variance for models with fixed parameters.
- Function
predict
can now return standard errors and prediction intervals. - Upgraded Roxygen2 to version 7.3.2.
- Updated vignette after review from the Journal of Statistical Software.
- Fixed again the
x
-axis of the plot.
- Fixed a bug in the
plot
when doses were not previously sorted. - Fixed a bug in the
x
-axis of theplot
when data contained zeros. - It is now possible to not plot data points by setting option
plot_data
toFALSE
(defaultTRUE
).
- Small fixes to unit tests because they were not passing on specific systems.
- Updated vignette with new simulation results.
It is now possible to fit models using either the log-dose or the dose scale.
To accommodate this extension it was necessary to change the default model parameterization, which now follows that of the Emax model (Macdougall, 2006).
Briefly, the 5-parameter logistic function is now defined as
alpha + delta / (1 + nu * exp(-eta * (x - phi)))^(1 / nu)
Parameter alpha
is the value of the function when x
approaches -Inf
.
Parameter delta
is the (signed) height of the curve.
Parameter eta > 0
represents the steepness (growth rate) of the curve.
Parameter phi
is related to the mid-value of the function.
Parameter nu
affects near which asymptote maximum growth occurs.
Similarly, the newly implemented log-logistic function (when x >= 0
) is
defined as
alpha + delta * (x^eta / (x^eta + nu * phi^eta))^(1 / nu)
Check the vignette (vignette("drda", package = "drda")
) or the help page (help(drda)
) to know more about the available models.
Here is a change log from previous version:
- Change parameterization to follow that of the Emax model.
- Implement the log-logistic family of models.
- Improve initialization algorithm to be more efficient and (hopefully) robust.
- Exported functions for evaluating theoretical gradient and Hessian of each implemented model.
- Implement the
effective_dose
function for estimating effective doses. - Added examples to help pages.
- Many minor bug fixes (too many to list them all).
First public release.