Do disaggregated CPI data improve the accuracy of inflation forecasts?☆

a r t i c l e i n f o In this paper, we evaluate the role of using consumer price index (CPI) disaggregated data to improve the accuracy of inflation forecasts. Our forecasting approach is based on extracting the factors from the subcompo-nents of the CPI at the highest degree of disaggregation. The data set contains 54 macroeconomic series and 243 CPI subcomponents from 1992 to 2009 for Mexico. We find that the factor models that include disaggre-gated data outperform the benchmark autoregressive model and the factor models containing alternative groups of macroeconomic variables. We provide evidence that using disaggregated price data improves forecasting performance. The forecasts of the factor models that extract the information from the CPI disaggre-gated data are as accurate as the forecasts from the survey of experts. Inflation forecasts play an important role in the effective implementation of an inflation targeting regime (Svensson, 1997). Moreover , many economic decisions, whether made by policymakers, firms, investors, or consumers, are often based on inflation forecasts. The accuracy of these forecasts can consequently have important repercussions on the economy. One possible way to improve the accuracy of inflation forecasts is to employ the information contained in the consumer price index (CPI) disaggregated data. However, traditional models such as vector autore-gressions would require a large number of parameters to estimate. Most previous literature about predictions of economic aggregates based on disaggregated information has focused on forecasting the component indices and aggregating such forecasts. Some studies in this line include, for example, Fair and Shiller (1990) for United States GNP; Zellner and Tobias (2000) for industrialized countries' GDP growth; Hubrich (2005) for forecasting euro area inflation; Marcellino et al. (2003) for disaggre-gation across euro countries; Bruneau et al. (2007) for French inflation; Moser et al. (2007) for Austrian inflation; Duarte and Rua (2007) for Portuguese inflation; and Capistran et al. (2010) for Mexican inflation. Among the papers related to inflation forecasting, the majority of them have used a rather low level of disaggregation, with the exception of Duarte and Rua (2007), who considered almost 60 subcomponents. Some other papers, such as Barhoumi et al. (2010), have investigated whether it is more appropriate to use aggregate or disaggregate data to extract the factors using a general definition of disaggregated data. That is, those authors use a broader set of macroeconomic variables rather than disaggregates of the variable to forecast. In …