pull down to refresh

Hmmm.... the timing of this change and the pressure to lower inflation seems very suspect. The last thing one does statistically when there is a lot of attention is change up the methodology on how metrics are arrived at unless there is a very good justification for it.

Typically, you’d want to run the old methodology in parallel with the new one for a while, so you get a comparison and see the trends in each.

reply

Correct. Redundancy is your friend here. However, having been on a number of transition projects in agencies, management never sees the value in that. They only see duplicative costs. So they jump off the branch wholesale without knowing how they will land forcing full adoption, even if its quickly seen as faulty.

reply

I’ve mostly seen the opposite: complete unwillingness to look at better methods because they aren’t directly comparable to the historical series.

reply

ugh. been there too, especially with entrenched types who resist change because they like things the way they are and change would mean power-sharing.

reply

transparency matters here. people are naturally going to question revisions that make inflation look better.

reply
5 sats \ 0 replies \ @Troser 20 Jul -30 sats

changing how you measure inflation doesn't change what people are paying at the grocery store..