Estimation of inflation with higher frequency and its usefulness for nowcasting

Recently the release of Wholesale Price Index in India has shifted to monthly from weekly frequency. Non-availability of the more frequently available WPI could lead to uncertainty and confusion particularly in an inflationary environment like that in India at present and such developing countries. However, the Primary Articles and FPLL sub-group Indices are released on weekly information. This paper highlights a framework formalizing the updating of the nowcast, particularly in respect of inflation as various data are released throughout a month and that can be used to evaluate the marginal impact of new data releases on the precision of the now/forecast as well as the marginal contribution of different groups of variables. Two methods are attempted to interpolate the weekly WPI (Unobserved) using the available information and the past data. The sate space model based on Kalman filter is found to be superior to the pure empirical method based on past data.