Advances and Applications in Statistics
Volume 51, Issue 5, Pages 367 - 388
(November 2017) http://dx.doi.org/10.17654/AS051050367 |
|
COMPARATIVE ARIMA MODELS FOR AGE-SPECIFIC FERTILITY RATES
Hend Auda and Hend Attia
|
Abstract: Getting the forecasted values of the age-specific fertility rates (ASFR) is very important to prepare population projections. Either population projections are deterministic or probabilistic. The question is how to get the forecasted values of ASFR. One possible technique is obtaining ASFR through the United Nations model age patterns of fertility (UN age patterns). This technique is the common technique in most studies for Egypt. Another possible suggested technique is modeling ASFR themselves. This paper mainly examines the question of which one of the two previous techniques would provide us with more accurate results. This question is considered empirically based on integrated autoregressive moving average (ARIMA) models using annual data on Egypt during the period from 1966 to 2010. This comparative study is conducted between: (a) univariate ARIMA model for the total fertility rate then used the UN age patterns; and (b) multivariate ARIMA model to the vector of ASFR. The results of the accuracy measures indicate that multivariate ARIMA model has the ability to clearly represent ASFR. Consequently, multivariate ARIMA model is used for forecasting ASFR in Egypt until year 2030. |
Keywords and phrases: total fertility rate, univariate ARIMA models, the UN age patterns, age-specific fertility rates, multivariate ARIMA models. |
|
Number of Downloads: 454 | Number of Views: 1541 |
|