Modeling postgraduate students flow

The average time taken to complete a course and the completion rate in higher education vary among students at a university according to their field of study, age and mode of study. Moreover, without any quantitative tools in predicting student admissions into the university, its management will encounter problems in setting up their strategic planning and allocating their resources. Hence, this study aims to predict the number of student enrolments each semester and understand the flow of students in a system, such as the mean time they spend in their postgraduate programs, and the estimated probability of graduating as well as dropping out from their studies.A data of full-time and part-time postgraduate students at Universiti Utara Malaysia (UUM) was chosen for this study.Previous works have been discussed on several methods used in an enrolment projection, and the most suitable method to be used in this study is the Markov Chain Model.The validity of the model is evaluated by comparing it with the historical data using mean absolute percent errors (MAPE).The result shows that the Markov Chain Model excels in making the enrolment projection, and the absorbing state used can analyse the system in further depth