How the co-integration analysis can help in mortality forecasting

Abstract

The method of mortality forecasting proposed in 1992 by Lee and Carter describes a time series of age-specific log-mortality rates as a sum of an independent of time age-specific component and a bilinear term in which one of the component is a time-varying parameter reflecting general change in mortality and the second one is an age-specific factor. Such a rigid model structure implies that on average the mortality improvements for different age groups should be proportional, regardless the calendar period. In this paper we investigate whether the mortality data for England and Wales follow this property or not. We perform the analysis by applying the concept of the Engle and Granger co-integration to the time series of log-mortality rates. We investigate the goodness of fit of the predictions to the historical data. We find that a lack of co-integration indeed can cause some problems in performance of the model. In the last section we propose several opportunities to omit the pitfalls.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.