Feature Request: dr4pl() to implement additional parameters similar to drc::drm()

Issue #5 new
Ghislain B created an issue

Hi,

DR4PL is a great module. I think it would be great, it would be great if dr4pl could implement a few features implemented in drc:drm() although I appreciate this might be a lot of work.

I am particularly interested in being able to constrain the curve fitting (I did not find any ways to do this in its current state). If these parameters could be implemented that would be excellent:

*lowerl: a numeric vector of lower limits for all parameters in the model (the default corresponds to minus infinity for all parameters).

*upperl: a numeric vector of upper limits for all parameters in the model (the default corresponds to plus infinity for all parameters).

*fct: In particular, supporting these functions: LL.1(), LL.2(), LL.3() and LL.4() (this could be extended in the future)

*curveid: a numeric vector or factor containing the grouping of the data.

*pmodels: a data frame with many columns as there are parameters in the non-linear function. Or a list containing a formula for each parameter in the nonlinear function


It would be great if the objects returned by dr4pl, could "extend" drc object (or at least implement most of the elements from a drc object). This could perhaps be done in a similar fashion as data.table, extends data.frame This would allow to easily convert some code from DRC to DR4PL, and use some of the statistical method devised for DRC class objects. I realize this might be a whole lot of work, but I think it would help with the adoption of this great package!

Comments (3)

  1. Justin Landis

    Hello Ghislain,

    Currently all development has come to a freeze and we do not expect to update the package until June of 2018. In the meantime we would love to know more about the specific statistical methods you would like implemented. Additionally, please initiate a pull request if you develop any feature you feel should exist in dr4pl. We would love to look over any suggestions to improve the package's usefulness.

  2. Log in to comment