Testing and dating of structural changes in practice
Thereby, we focus on male mortality trends, where the impact of the smoking epidemic already peaked in most countries, while it is generally still increasing among females, so that structural changes due to smoking might occur more likely in the future.
If smoking indeed explained most of the breaks in males, projection models including information on smoking should become a standard approach in mortality forecasting.
If your browser does not accept cookies, you cannot view this site.
In the presence of such nonlinearity, past trends do not provide a solid basis for extrapolation and the resulting projections become more volatile and particularly sensitive to the selection of the historical period (Peters et al. To develop better projection models, knowledge on the underlying determinants responsible for the structural breaks is required.
Taking the tobacco epidemic into account should become standard procedure in mortality forecasts to avoid a misleading extrapolation of trends.
Nevertheless, more research is needed to identify additional factors, such as health-care policies and innovations in medical treatment, to explain the remaining structural changes.).
Different parts of the UK were analyzed separately because large spatial variation has been reported with respect to mortality trends in the UK different areas (Murray ).
We did not include smaller countries (such as Iceland and Luxembourg), where irregularities in time trends might be due to the small number of events rather than to external influences.