Closed gracezhihuizhao closed 1 year ago
@gracezhihuizhao , I was thinking about the "decreasing" curve and biphasic curve situations and considered to see if there are others that have an analogous situation/problem. Here is something that I found, with this being the case I think we may want to consider all "areas" positive. Thus, when curve decreases from 0 then instead of a negative value we will still want to have a positive "AUC" (or really area above the curve), and for curves that pass the x-axis then we should add the absolute value of the area sections. https://math.stackexchange.com/questions/3223633/area-under-the-curve-integrals-antiderivatives Let me know what your thoughts are. I think this approach may be the most appropriate for the application.
@Kelly-Carstens-EPA and @kpaulfriedman feel free to chime in if you have any thoughts.
@sedavid01 I agree adding the absolute value of the AUC components from a biphasic curve probably makes the most sense. I think of AUC as an indicator of overall effect size.
Going to move the case for curves crossing the x-axis to an extension/enhancement ticket to keep this one moving forward.
-implemented get_AUC function -included a simple vignette showing how the function works