An autoregressive time series model of measles propagation: a case study of the Kumasi District

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
In 1985, the Government of Ghana assisted by the World Bank and other organizations, initiated a programme of mass vaccination aimed at eradicating six childhood killer diseases namely, measles, poliomyelitis, whooping cough, diphtheria, tuberculosis and tetanus. In this study we attempt to assess the effectiveness of this intervention for the particular case of measles incidence in the Kumasi District. Using the Box-Jenkins method we obtain autoregressive (AR (2)) models of the pre intervention data, post intervention data and the entire interrupted times series data. Further, using interrupted time series analysis, we are able to show that the parameters before and after the intervention were significantly different at the 5% level of significance. In particular, the intercepts of linear trend lines were 27.56 and 29.04 cases per month before and after the intervention respectively. The slopes were —0.084 and -0.130 cases per month before and after the intervention respectively. The methods used in this study can clearly be used to investigate the effectiveness of any intervention aimed at controlling some time dependent natural or socio-economic phenomena.
A thesis submitted to the Board of Postgraduate Studies Kwame Nkrumah University Of Science And Technology, Kumasi, Ghana in partial fulfilment of the requirements for the award of the degree of Master of Science (M.Sc)in Mathematics, 2001