I utilise an engineering tool to investigate periodicity for all cause death within England & Wales by date of death for the period 1st June 2014 – 31st July 2023
Can you explain this in ~layman's terms...'there is a notable absence of 30-day periodicity'? Does this mean that every 30 days during the period studied (during a subset of the period studied?) there are significantly fewer deaths than expected?
Not fewer deaths but less harmonic content than expected. If the death count dropped to zero at the end of every month this would, in fact, generate a sizeable harmonic spike at f = 0.033, as would a sudden surge in death at the end of every month. These analyses/spikes have nothing to do with absolute numbers and everything to so with the harmonic patterns hidden within those numbers. They tell us if there is a pattern to deaths over time, regardless of whether these are increasing or decreasing. An analogy is a tuning fork, which sounds because the armatures move inward (=fewer deaths) then outward (=more deaths). We are not concerned with the position of the armatures (count of deaths) but we are concerned about the frequency of the sound it makes if we are trying to tune our instrument to concert pitch!
Not in terms of the death count. A spike goes up only because there is harmonic content at that frequency. A spike goes down if there isn't any harmonic content. In sum these spikes are revealing frequencies that are present within the time-dependent data series.
Can you explain this in ~layman's terms...'there is a notable absence of 30-day periodicity'? Does this mean that every 30 days during the period studied (during a subset of the period studied?) there are significantly fewer deaths than expected?
Not fewer deaths but less harmonic content than expected. If the death count dropped to zero at the end of every month this would, in fact, generate a sizeable harmonic spike at f = 0.033, as would a sudden surge in death at the end of every month. These analyses/spikes have nothing to do with absolute numbers and everything to so with the harmonic patterns hidden within those numbers. They tell us if there is a pattern to deaths over time, regardless of whether these are increasing or decreasing. An analogy is a tuning fork, which sounds because the armatures move inward (=fewer deaths) then outward (=more deaths). We are not concerned with the position of the armatures (count of deaths) but we are concerned about the frequency of the sound it makes if we are trying to tune our instrument to concert pitch!
Is there a difference in interpretation if the spike goes up vs. down?
Not in terms of the death count. A spike goes up only because there is harmonic content at that frequency. A spike goes down if there isn't any harmonic content. In sum these spikes are revealing frequencies that are present within the time-dependent data series.
Is it possible for me to message you directly? Would be interested in your thoughts on a dataset I've been looking at..
I'm afraid my time commitment to existing and planned studies is 100% for the foreseeable future.