Nowcasting inflation using high frequency data

This paper proposes a methodology to nowcast and forecast inflation using data with sampling frequency higher than monthly. The nowcasting literature has been focused on GDP, typically using monthly indicators in order to produce an accurate estimate for the current and next quarter. This paper exploits data with weekly and daily frequency in order to produce more accurate estimates of inflation for the current and followings months. In particular, this paper uses the Weekly Oil Bulletin Price Statistics for the euro area, the Weekly Retail Gasoline and Diesel Prices for the US and daily World Market Prices of Raw Materials. The data are modeled as a trading day frequency factor model with missing observations in a state space representation. For the estimation we adopt the methodology exposed in Banbura and Modugno (2010). In contrast to other existing approaches, the methodology used in this paper has the advantage of modeling all data within a unified single framework that, nevertheless, allows one to produce forecasts of all variables involved. This offers the advantage of disentangling a model-based measure of ”news” from each data release and subsequently to assess its impact on the forecast revision. The paper provides an illustrative example of this procedure. Overall, the results show that these data improve forecast accuracy over models that exploit data available only at monthly frequency for both countries.