Free virtual event, Friday 7th October 2022, 14:00-17:00 UK time
Background
Existing fertility projection models tend to use data on total births rather than disaggregating by birth order. However, there are many reasons to incorporate parity information, not least that it allows closer control of the population at risk of each birth order. A parity-specific approach could therefore lead to more reliable fertility projections and enable more tailored policy solutions. This is the second in a series of two training workshops, the first on “Fertility Analysis by Birth Order”. Attendance at the first workshop is not a formal prerequisite, but it would provide useful background knowledge.
Objectives
The workshop will summarise existing fertility projection methods and will then introduce data sources and a method that can be used to obtain Bayesian parity-specific fertility projections. This method will be demonstrated in the practical session.
Audience
The workshop is aimed at national and local government statisticians, particularly those who contribute to population projections; social and demographic researchers; and others interested in the application of sophisticated statistical methods to demographic forecasting.
Content
Fertility projections – theory (45 min)
- Existing fertility projection methods
- A proposed method for obtaining parity-specific fertility projections
Fertility projections – practical session (2 h)
- Projection of fertility at the aggregate level under the proposed method
- Projection of fertility at the parity-specific level under the proposed method
Software
Participants should have a basic knowledge of R (be able to load and run R code), and preferably RStudio (see ModernDive for an entry-level tutorial and installation instructions). Demos will use the R package ‘RStan’ (installation instructions available here); prior experience with RStan is not required. Participants will need to register for free with the Human Fertility Database prior to the workshop. Participants can simply watch the demos, but ideally should download the software to be able to run the analyses on their own machines. The required software is free to download and straightforward to install.
Instructors
Dr Joanne Ellison, Prof Jakub Bijak, Prof Erengul Dodd (University of Southampton, UK)
Background reading
Bohk-Ewald, C., Li, P., & Myrskylä, M. (2018). Forecast accuracy hardly improves with method complexity when completing cohort fertility. Proceedings of the National Academy of Sciences, 115(37), 9187–9192. https://doi.org/10.1073/pnas.1722364115.
Ellison, J. (2022). Forecasting fertility. In Petropoulos, F., et al. Forecasting: theory and practice. International Journal of Forecasting, 38(3), 705–871. https://doi.org/10.1016/j.ijforecast.2021.11.001.
Acknowledgements
This workshop is funded by the ESRC FertilityTrends project (grant number ES/S009477/1).