Exploring the Methods of Cointegration Procedures in Dynamics of HIV/AIDS and Related Infections

Abstract

Introduction: Cointegration has become an important property in contemporary time series analysis. Time series often have trends - either deterministic or stochastic. Material and Methods: This research work seeks to determine the inherent long run relationship between the human immunodeficiency virus (HIV) and other diseases and the circumstances when it is reasonable to expect that two or more diseases may be cointegrated. That is, if at least one of the processes is driving the other and if the diseases are being driven by the same underlying process. Result: The data (Prevalence of HIV/Acquired Immunodeficiency Syndrome [AIDS] Incidences) modeled in this study were obtained from the National Agency for the Control of AIDS. The stationarity characteristics of the study variables were investigated using Augmented Dickey-Fuller test, the long-run relationship between HIV and other two diseases was determined using Engle-Granger cointegration, Phillips-Ouliaris, and Johansen testing procedure while the Granger causality test was also performed to determine the short run relationship of the variables. Conclusion: Results showed that the series are integrated of order two; HIV, Hepatitis, and TB are found to be strongly and significantly positively correlated, the data series are considered to be stationary after the second differences. Furthermore, the Granger causality tests show that HIV “Granger causes” Tuberculosis and Hepatitis in Nigeria. However, Phillips-Ouliaris Test for Cointegration Determines the strongest cointegration level among HIV/AIDS and other infections. Hence, it is the most robust test for testing cointegration between HIV and tuberculosis, and HIV and Hepatitis disease.