One of the barriers
of more widely employing the proven-useful ProFamy extended cohort-component method is that the sex-age-specific occurrence/ exposures rates of
marriage/union formation and dissolution and age-parity-specific occurrence/exposures rates of marital and non-marital fertility, which are needed for household projections, are not yet readily available for many countries. However, the age-sex-specific occurrence/exposures rates (defined as number of events that occurred in the age interval divided by the number of person-years at risk of experiencing the event) can in fact be relatively easily estimated using standard statistical software and fertility and marriage histories data collected in conventional national fertility/family surveys in most countries. Once the age-race-sex-specific standard schedules at the national level are prepared, they can be readily employed as model standard schedules for projections
at the sub-national and county levels. The model standard schedules can also be used for household projections in other countries which have similar demographic patterns but do not yet have the needed data to estimate the age-specific schedules. This is similar to widely practiced applications of model life tables (e.g., Coale, Demeny, and Vaughn, 1983; United Nations, 1982), the Brass logit relational life table model (e.g. Murray et al., 2003), and other parameterized models (e.g. Coale and Trussell, 1974; Rogers, 1986) in population projections and estimations.
Similar to use of model life tables together with estimated life expectancies in population projections, sex-age-specific model standard schedules should be accompanied by the summary indices for household projections. This is because the numerous age-sex-specific rates cannot concisely represent the quantum and tempo
of the demographic processes, and thus summary demographic measures (e.g., TFR, mean age of fertility, general marriage and divorce rates) are needed. Furthermore, as Keyfitz (1972) points out, demographic projections based on trend extrapolation of each age-sex-specific rate can result in an excessive concession to flexibility and readily produce erratic results. Accordingly, once the model standard schedules are prepared, analysts can concentrate on projecting future demographic summary parameters. This can be done, for example, using conventional time series analysis by statistical software (e.g., SAS, SPSS, or STATA) or expert opinion. Time series data on other related socio-economic covariates (e.g., average income, education, urbanization, etc.) can also be used to project future demographic summary parameters.
A notable example is that local governmental office employed ProFamy method/software, the U.S. national race-sex-age-specific model standard schedules and the demographic summary parameters at the county-level to successfully project households and living arrangements for the six counties of Southern California since 2009, with projections renewed every two years. The six counties’ governments have effectively used these detailed biennial projections for their socio-economic planning, budget allocations, and policy analyses on housing, traffic, energy consumption, elderly care and other home-base social services (Feng, Choi et al., 2018).
Numerous studies have demonstrated that that model standard schedules and a few summary parameters offer an efficient and realistic approach for demographic projections (Brass, 1978; Booth, 1984; Paget and Timaeus, 1994; Zeng et al., 1994). The theoretical foundation of this practice is that the demographic summary parameters are crucial to determine changes in level and age pattern of the age-specific rates which affect the projections. At the same time, the projection and estimation results typically are not highly sensitive to the sex-age-specific model standard schedules, as long as the possible changes in the level of occurrences and timing of the demographic events are properly measured by the relevant summary parameters. Thus, the sex-age-specific model standard schedules using national survey data can be readily used for household projections at the sub-national and county levels as empirically tested in Zeng et al. (2006; 2013), or even be used for projections or estimations in other countries with similar demographic patterns, as corroborated in Zeng et al. (2000).