Forecasting inflation in G-7 countries: an application of artificial neural network

Abstract

Purpose

The paper aims to evaluate different artificial neural network models and to suggest a suitable model for forecasting inflation in G-7 countries.

Design/methodology/approach

The study applies different combinations of neural networks with hyperbolic tangent function using backpropagation learning with the steepest gradient descent technique to monthly data on Consumer Price Index (a measure of inflation) of the USA, the UK, France, Germany, Italy, Japan and Canada.

Findings

Predictions of inflation based on the Consumer Price Index for all the seven countries divulged that it is expected that the rate of inflation will decline marginally in the near future.

Practical implications

The results proposed in this study will be a benchmark for policy-makers, economists and practitioners to forecast inflation and design policies accordingly.

Originality/value

The paper’s findings provide strong evidence for policy-makers that while constructing models for forecasting inflation, the suggested models can be used to track the future rates of inflation and, further, they can apply that model in framing policies.

Citation

Gupta, S. and Kashyap, S. (2015), "Forecasting inflation in G-7 countries: an application of artificial neural network", Foresight, Vol. 17 No. 1, pp. 63-73. https://doi.org/10.1108/FS-09-2013-0045