Premium
Measuring Core Inflation in S outh A frica
Author(s) -
Plessis Stan,
Rand Gideon,
Kotzé Kevin
Publication year - 2015
Publication title -
south african journal of economics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.502
H-Index - 31
eISSN - 1813-6982
pISSN - 0038-2280
DOI - 10.1111/saje.12090
Subject(s) - core inflation , inflation (cosmology) , economics , core (optical fiber) , econometrics , measure (data warehouse) , price index , wavelet , dynamic factor , sample (material) , variable (mathematics) , macroeconomics , monetary economics , inflation targeting , monetary policy , computer science , mathematics , telecommunications , chemistry , physics , chromatography , database , artificial intelligence , theoretical physics , mathematical analysis
Abstract Measures of core inflation convey critical information about an economy. They have a direct effect on the policymaking process, particularly in inflation‐targeting countries, and are utilised in forecasting and modelling exercises. In South Africa, the price indices on which inflation is based have been subject to important structural breaks following changes to the underlying basket of goods and the methodology for constructing price indices. This paper seeks to identify a consistent measure of core inflation for South Africa using trimmed means estimates, measures that exclude changes in food and energy prices, dynamic factor models, and wavelet decompositions. After considering the forecasting ability of these measures, which provide an indication of expected second‐round inflationary effects, traditional in‐sample criteria were used for further comparative purposes. The results suggest that wavelet decompositions provide a useful measure of this critical variable.