search

UT-Covid / episimlab

Framework for development of epidemiological models
https://ut-covid.github.io/episimlab/
BSD 3-Clause "New" or "Revised" License
3 stars 1 forks source link

Residual error incorrectly calculated #7

Closed ethho closed 3 years ago

ethho commented 3 years ago

Currently calculating residual error as the abs(sum(actual - pred)): https://github.com/eho-tacc/episimlab/blob/e888ae7d7f1d1773c271c53320a7ea8a341b40d4/episimlab/fit/llsq.py#L109

This is incorrect; nominal time-scaled vector is the format we want.

I think vector residual error is more informative, using the nominal (not absolute) value. Usually you check to see that the residuals are normally distributed with a mean of 0, and that if you plot residual vs time you don't see constriction of residual error in any one spot