Thanks, that’s a very good point — I am using the unweighted median.
It is important to explain why. Inflation in any individual country can be thought of as the sum of an underlying component that is correlated internationally and a country-specific, erratic (noise) component. It is this latter element that makes inflation difficult to interpret, so the aim is to reduce its influence.
There are nine countries in my sample. If I computed a GDP-weighted median, the United States would probably account for close to 50% of the total weight. In that case, idiosyncratic US noise would still be important for the median. By using an unweighted median, I can reduce its importance.
Thanks Stefan. Would you expect the US correlation with the ROW weaken overtime due to tariffs ?
Hi Priyanka. That is a good point -- of course, we would! Thanks.
Thank you Stefan for this analysis. Just to clarify: the median you’re showing is unweighted?
Thanks, that’s a very good point — I am using the unweighted median.
It is important to explain why. Inflation in any individual country can be thought of as the sum of an underlying component that is correlated internationally and a country-specific, erratic (noise) component. It is this latter element that makes inflation difficult to interpret, so the aim is to reduce its influence.
There are nine countries in my sample. If I computed a GDP-weighted median, the United States would probably account for close to 50% of the total weight. In that case, idiosyncratic US noise would still be important for the median. By using an unweighted median, I can reduce its importance.
Understood, thank you.
Important topic
https://organon.substack.com/s/economics/archive?sort=new