Time series analysis of fuel prices in Uganda using an ARIMA model from 2015 to 2018
Mugagga, Mark Treasure
MetadataShow full item record
A time series is a sequence of data points, typically measured at uniform time intervals. Examples occur in a variety of fields ranging from economics to engineering, and methods of analyzing time series constitute an important part of Statistics. Time series analysis comprises methods for analyzing time series data in order to extract meaningful characteristics of the data and forecast future values. The Autoregressive Integrated Moving Average (ARIMA) models, or Box-Jenkins methodology, are a class of linear models that are capable of representing stationary as well as non-stationary time series. The ARIMA models incorporate three key components: autoregressive (AR), differencing (I) and moving average (MA). The AR component considers a relationship between past and present fuel prices, while the MA component models the dependency on the error terms. The differencing component aids in addressing non-stationarity ensuring that data exhibits a stable behavior over time. In reference to the oil and gas (fuel) there is no local production as of late and it imports all its products from overseas about 95% routed through Kenya and 5% route through Tanzania. The trend of fuel prices in Uganda is influenced by a number of factors such as; forces of demand and supply, taxes, political events, weather conditions as well as exchange rates. The aim of the model is to understand the patterns and trends of fuel prices using monthly data from periods of 2015 to 2018 and develop a model that can accurately predict future prices which plays a critical role in various sectors such as transportation, energy, economic forecasting, trade and commerce. The forecasted fuel price values are compared to actual values to evaluate the model’s predictive performance. The study goes on to explore the impact of exogenous variables such as crude oil prices, economic indicators and geopolitical events on fuel price dynamics incorporating them into the ARIMA models as necessary. Furthermore the findings of the research contribute to a better understanding of fuel price behavior and provide valuable insights for decision-makers in sectors affected by fuel price fluctuations. The developed ARIMA models offer accurate and reliable forecasts, aiding in strategic planning, risk management and policy formulation related to fuel pricing.