Sujata Kar
College of Management and Economic Studies, University of Petroleum and Energy Studies, Energy Acres, Bidholi, Dehradun, Uttarakhand.
Abstract
The objective of this paper is to find out a measure of core inflation that could satisfy the major criteria suggested in the literature to judge how good a core inflation estimate is. Using WPI inflation for the period 1989:2 – 2005:12, the analysis shows that geometric exponential smoothing and weighted percentile are the most suitable statistical tools which can produce core inflation estimates that are efficient and have properties similar to WPI inflation. Further, while the core inflation estimates from geometric exponential smoothing can produce good forecasts of the benchmark, forecasts for the headline inflation is reasonable from weighted percentile for certain specific periods at a regular interval.
JEL
Classification:
E31, E37, E58
Keywords : Core Inflation; Statistical Tools; Out-of-sample Forecasts