search

BiologicalRecordsCentre / BRCindicators

An R package for creating indicators from trends data
4 stars 11 forks source link

rescale_Yr doesn't behave as expected #67

Closed drnickisaac closed 4 years ago

drnickisaac commented 4 years ago

@galinajonsson noticed unexpected behaviour. rescaleYr treats the uncertainty as expected but does not scale the overall indicator to the baseline value in this year. Should be easy to fix.

AugustT commented 4 years ago

is this internal to the bma function?

drnickisaac commented 4 years ago

Yes, rescaleYr is an argument to bma(). This is what I will explore and fix. But it's possible there are rescaling options elsewhere in the package.

drnickisaac commented 4 years ago

I have explored the behaviour of rescaleYr in bma() and it behaves as I expected, at least when errorY1 = FALSE. When errorY1 = TRUE the only noticeable effect is to shift the line up or down. So I imagine that @galinajonsson used a different function for rescaling. I'll close this issue now: there remains a question about which options we make available, and how they are explained/expressed. But that's more general than this issue, and probably falls under wider issues around harmonization and documentation.