As you can see (above), this data is highly seasonal. So I have applied a seasonal decomposition filter to get the next plot. The key message is that the trend is down, but the month on month seasonally adjusted result is up. The seasonally adjusted series is still pretty noisy, so this latest move up in this series might not be meaningful.
Just for the fun of it I have also attached a couple of the seasonal adjustment diagnostic plots. The first of these plots summarises the original data, the seasonal component, the trend component and the unexplained remainder. The second plot looks at the seasonal component for each month. On average in January, ASIC processed some 260 fewer applications than the monthly average for the entire period. Over time, the size of the January downturn has been getting bigger. December and April are the other low months. March is typically the peak workload month.
Around the states we have the following seasonally adjusted outcomes:
No comments:
Post a Comment