期刊
INTERNATIONAL JOURNAL OF FORECASTING
卷 40, 期 1, 页码 160-183出版社
ELSEVIER
DOI: 10.1016/j.ijforecast.2023.01.008
关键词
Global inflation; Common factors; Forecasting; Inflation spillovers; Machine learning; Variable selection
This article examines whether inflation is a global phenomenon for European emerging market economies and finds that a global inflation factor explains more than half of the variability in national inflation rates. The study shows that forecasting models that include global inflation factors generally provide more accurate predictions of national headline inflation rates compared to Phillips curve-type models and models with local inflation factors. The results are robust to considerations of sparsity and non-linearity in the factor forecasting models. The article also offers insights into the country-level characteristics that increase the importance of global factors in domestic inflation.
We consider whether inflation is a 'global phenomenon' for European emerging market economies, as has been claimed for advanced or high-income countries. We find that a global inflation factor accounts for more than half of the variance in the national inflation rates, and show that forecasting models of national headline inflation rates that include global inflation factors generally produce more accurate path forecasts than Phillips curve-type models and models with local inflation factors. Our results are qualitatively unaffected by allowing for sparsity and non-linearity in the factor forecasting models. We also provide some insight as to why global factors are an important determinant of domestic inflation, by considering the country-level characteristics that tend to increase the importance of global factors for domestic inflation. (c) 2023 The Author(s). Published by Elsevier B.V. on behalf of International Institute of Forecasters. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据