Inflation dilemma

For Central Banks (CBs), the problem is that as economic indicators change direction, it remains unclear for the first few months after the peak or trough that they have changed direction.  You need enough data that the underlying trend overwhelms random fluctuations.  The latest release of Australia's new monthly CPI exemplifies this.  It seems probable that inflation is peaking.  But is it peaking at the current level, or is it about to start falling?  How many months/quarters of data will the Reserve Bank of Australia need before they can be certain that they have done enough tightening to break inflation's back?  After all, it did look as if, in 2021, inflation had peaked, only for it to surge again in 2022.

This isn't a dilemma which faces only the RBA.  The Fed, the ECB (European Central Bank), and other central banks face it too.  And CB's response is usually to keep policy too tight for too long, or to not raise rates early enough to head off higher inflation.  CB discount rates tend to lag the cycle.  It's not because they're stupid, or inept, though many who do not have their job or their responsibility think that they are.  It's because economic time series have big random components, and it takes time for the underlying trends to become clear.  

To mitigate the problem of random fluctuations, I tend to extreme-adjust my time series, and to use an average of more than one time series, on the argument that when one time series had a random uptick, it's probable that another has a random down tick, so the average has a smaller random error.   Also, I like to analyse leading indicators because they give clear warning of impending changes in direction.  

Leading indicators currently suggest that economies are already being impacted by rising rates, and that CBs should be very cautious about further rate rises, because the economic response lags the change in interest rates by many months.  Rate rises now are likely to worsen the downturn in economies.