ИСТИНА

Horiuchi & Coale (1982) have suggested a simple formula to improve the estimates of the expectation of life at old age, a formula that was questioned as unrealistic by Mitra (1984; 1985), whose alternative approach was in return criticized for being prone to biases due to age exaggeration (Coale 1985). The discussion between the researchers have not been resolved. This lack of agreement and the simplifying assumptions behind both approaches prevented their integration in the practical toolkit of demographic analysis. This paper shows that both methods are useful in drastically reducing the estimation errors in life expectancy. The disputes were largely due to inconsistent inputs rather than methodological contradictions. Furthermore, the methods produce by far better results as compared to the popular alternative, the extrapolation of the death rates. In combination the two alternative methods may even lead to a better outcome.