These tools will no longer be maintained as of December 31, 2024. Archived website can be found here. PubMed4Hh GitHub repository can be found here. Contact NLM Customer Service if you have questions.


PUBMED FOR HANDHELDS

Search MEDLINE/PubMed


  • Title: Analysis and forecasting infant mortality rate (IMR) in Egypt until year 2000.
    Author: Hussein MA.
    Journal: Egypt Popul Fam Plann Rev; 1991 Dec; 25(2):32-46. PubMed ID: 12288769.
    Abstract:
    Two models, one with the natural logarithmic transformation of infant mortality time series and the other with successive differences, were used to provide infant mortality projections for the period 1983-2000 in Egypt. Data were obtained from CAPMAS and UNICEF on the Egyptian infant mortality rate for the period 1947-82. The best model was determined by successive steps of model specification, estimation, and comparison. Plots of the data were provided for the original data for 1947-82, the degree of non-seasonal differencing, and a natural log transformation of the data. Plots were also provided of the sample autocorrelation function and the sample partial autocorrelation function for the original data, the degree of differences, and the natural logarithmic transformations. The preferred model was an autoregressive integrated moving average one for a first difference model (model 1) and a natural logarithmic model (model 2). Parameter estimates in model 2 were more significant and therefore preferred. Goodness of fit comparisons and comparisons of plots of sample autocorrelation functions for the errors with their probability limits showed both models to be adequate. The two models were used to forecast infant mortality between 1983 and 2000. Model 1 showed a faster decline in mortality than model 2: a decline of 44.4% compared to 25.9%. Model 2 results were preferred because of the known inaccuracies in infant mortality data and the initially sharp decline between 1984 and 1985, which was due to implementation of government health programs. Caution is urged in interpreting the data as precise.
    [Abstract] [Full Text] [Related] [New Search]