Application of K-Nearest Neighbours and Long-Short-Term Memory Models using Hidden Markov Model to Predict Inflation Rate and Transition Patterns in Nigeria

Abstract

It fluctuates in most countries in the world regardless of whether the countries are been developed, developing or underdeveloped. The widespread effects of inflation have a significant impact on most countries including Nigeria because it is a global phenomenon influenced by variety of factors such as economic growth and monetary policy. This paper proposes the application of K-Nearest Neighbours (KNN) and Long Short-Term Memory (LSTM) models using Hidden Markov Model (HMM) to predict the inflation rate and its transition patterns in Nigeria with Secondary data collected from National Bureau of Statistics (NBS) official website. Empirical analysis revealed that GDP per capita show a significant influence in inflation rate and contributes to inflation forecasting in Nigeria.