Multivariate Analytics on Economic Impact of Oil and Gas Price Swing: Nigerian Outlook

Abstract

Abstract Oil prices have fallen dramatically to the average range of USD $30 to $60 per barrel and are expected to stay in this average range through 2015 and beyond. Historically, crude oil producing countries seemed to be more concerned on the risks and uncertainties associated with fluctuations in crude oil price economic wise globally. This was evident both in times of rising and falling price, imposing serious challenges to them. Moreover, Natural gas also contributes to Nigeria’s GDP resulting from global industrialization and more swing from crude oil price. The implications of these swings significantly influence the economy of the country in question (Nigeria), shaping the narrative around it. Statistical methods are distinctive in their ability to provide insights into correlations where thresholds exist, building a predictive statistical data-driven model for analyzing the economic impact of swings in crude oil and gas prices with corresponding output has been of great interest in the oil and gas producing countries primarily because of its successful implementation in several industries. The analytics behind the predictive models are described through mathematical functions that rely on monthly and yearly averaged spot oil and gas prices and total oil and gas produced from 2015 to 2019 data range. This research work evaluates the extent to which these variations on crude oil and gas prices with respect to their production effects on Nigerian economy thereby impose a need for a well fitted analytical model for future forecasting based on statistics. The Statistical-analysis tool used to clean and examine the data sets were Multiple Linear Regression Analysis (MLRA) coupled with analysis of variance (ANOVA) where calculations took place in data analysis add-ins tool pack of Microsoft office Excel work sheets using the functions set out by the methodology. Thorough analysis of the results showed that some of the models build to be highly fitted and statistically significant, evident from R-squared values, adjusted R-Squared, standard errors, other hypothesis and statistical assumptions detailed in the results and analysis shows that the basic objective of this research work is achieved.