Michalski, Anatoli I.; Grigoriev, Pavel; Gorlishchev, Vasily P. (2018)
MPIDR Technical Report TR-2018-002. Rostock
Occasionally, it is necessary to split aggregated fertility data into a fine grid of ages. In this technical report, we present an application of the quadratic optimization (QO) method for splitting abridged fertility data. The QO approach allows us to split not only age-specific fertility rates, but also birth counts in cases in which population exposures are available. Unlike the existing methods, the QO algorithm simultaneously satisfies five criteria of the ‘ideal’ disaggregation method: 1) shape – the estimated fertility curves should be plausible and smooth; 2) fit – the predicted values should be as close as possible to the observed values; 3) non-negativity – only positive values should be returned for the birth-order data; 4) balance – the estimated five-year age group totals should match the input data; and 5) parity – the balance by parity must be maintained. The examples provided in this report demonstrate the simplicity, flexibility, and high degree of effectiveness of the proposed technique.