In the case of the consumer price index the importance of a seasonally adjusted measure lies in the possibility of promptly identifying turning points in inflation. The focus of this paper is the so-called "cost-of-living" index, the most used measure of Italian inflation. Our aim is to identify and test a model of its seasonal component, which is important in explaining the overall variability of the index. Most seasonality is actually spurious, being essentially linked to the fact that the data on a significant number of the items included in the cost-of-living index are collected quarterly. It is therefore very important to take this into account in order to adjust the cost-of-living index for seasonality; the type of seasonality involved must also be considered, since different methods need to be used depending on the characteristics of the series considered.
On the basis of this preliminary analysis, the method implemented in the paper to adjust the series for seasonality is the model-based procedure known as SEATS. The measure recommended for monitoring the performance of inflation is the quarterly growth rate of the seasonally adjusted index, for two main reasons: first, a significant price change is always considered for all the series (i.e., for both monthly and quarterly series included in the cost-of-living index); second, it allows us to smooth out irregularities in price changes.
