Modelling Liquified Petroleum Gas Prices in Nigeria Using Machine Learning Models

Abstract

The usage of Liquefied Petroleum Gas in households has been to increase in recent years. The energy used by households has proved difficult to forecast due to the nature of the independent variables. In recent years, deep learning models has been broadly utilized in the machine learning area to model time series data. Most noticeable is in the field of forecasting. In this work, Neural Network Autoregressive model (NNETAR), Naive forecasting and the Autoregressive Integrated Moving Average (ARIMA) models were used to model the price of Liquefied Petroleum Gas prices(LPG) of 37 states (including the FCT) in Nigeria, with input variables in the form of the price of refilling LPG for 12.5 kg from January 2016 to April 2019 covering a 1480 data points. The Mean Absolute Percentage Errors was used to measure the performance of the model. The result shows that Naive produced lower MAPE for more states compare to NNETAR and ARIMA models.