Background: Life expectancy at birth (LE) reflects the overall mortality of a population and it is also used as one of the indicators of the health status of population.
Objective: The main purpose of this study is to project LE for the population of male, female and both sexes of urban region in Bangladesh by using polynomial model.
Data and Methodology: For this purpose, the data is obtained from Statistical Year Book (SYB) of Bangladesh of 2012. Quasi Newton Method is used to build up these models using the Statistical software STATISTICA. Furthermore, t-test statistics, F-test statistics and cross validity prediction power (CVPP) are applied to identify the correctness of these model.
Results: It is found that LE for the population of urban region of Bangladesh follow simple linear regression model. Based on t-stats, coefficient of determination, shrinkage coefficients and F-stats these models are seemed to be well fitted. So, these models can be used for better prediction. Then, these LE are projected during 2012-2051 using the fitted time trend models. It is observed that the LE for the population of male, female and both sexes of urban region of Bangladesh in 2051 are 75, 84.1 and 78 years respectively.
Conclusion: Urban population of Bangladesh are going to be aged population dramatically. Proper planning for the allocation of public and private resources may be taken for the better lifestyle of urban population.