jhhughes256 / optinterval

Raw R code for manuscript: Hughes JH, Upton RN, Reuter SE, Phelps MA, Foster DJR. Optimising time samples for determining AUC of pharmacokinetic data using non-compartmental analysis.
0 stars 0 forks source link

Perfecting the boundaries on study_rdata #11

Closed jhhughes256 closed 7 years ago

jhhughes256 commented 7 years ago

Goal

To highlight issues with study_rdata so that the broad study design works well.

Description

As seen in #7 there is a lack of precision of the algorithm. This could be due to the randomised curves produced by study_rdata being unrealistic.

This will require a case by case analysis of outliers, to determine if the random data is at fault, or if its the algorithm at fault.

jhhughes256 commented 7 years ago

Closing Issue For Now

It is a bit early to be blaming the randomised data for my algorithms problems. Additionally, difficult data should be handled with appropriate use of error messages, not by giving up!