Inconsistencies in proportion target looking computation

peekbank / peekbank-paper

peekbank initial paper

GNU General Public License v3.0

2 stars 2 forks source link

Inconsistencies in proportion target looking computation #20

Open mzettersten opened 2 years ago

mzettersten commented 2 years ago

There are a couple of slight inconsistencies in how proportion target looking is computed in the two use-case examples in the BRM paper. Specifically:

whether proportion target looking is computed using mean() vs. sum()
whether NA is removed in computing the denominator

I think this should make no difference to the actual values that are computed. But it would probably be good to make this consistent.

mzettersten commented 2 years ago

I made an edit here to make these computations in the two use-cases consistent (e68b0b9216099600705f6eb7a2d2cca1da9fd6b9). Resulting plots look identical. Leaving this open to make sure we check in about this.