Given the importance of insight in future demographic trends, many statistical agencies routinely compute national population forecasts.They do so by means of the so-called cohort-component model, which has become the standard approach in population forecasting (National Research Council -NRC 2000; UNECE 2018).This model requires assumptions on future trends of fertility, mortality, and international migration.We will discuss this approach further in Sect.1.2.To make accurate demographic forecasts is both an art and a science, similar to predictions in other fields (Tetlock and Gardner 2016).The scientific part is in the model, and in the fine mathematical and statistical details of the computations.However, to formulate reliable assumptions for the future course of fertility, mortality and migration is an art, largely.Most of the research on demographic forecasting aims at increasing the scientific part, and reducing the impact of selecting the right assumptions -the "art part" in population forecasting."The quest for knowledge about the future has moved from the supernatural towards the scientific" (Willekens 1990, 9).One way to achieve this aim is to formulate explicit models for fertility, mortality and migration.In that case, one attempts to find a model that describes the historical development of these components of change accurately enough.The model may be an explanatory model with exogenous variables, or a purely statistical (e.g.time series) model.In either case, the model is used to extrapolate the components into the future, and next their future values are used as inputs for the cohort-component model.The primary aim of this book is to sketch new developments in the scientific part of demographic forecasting.It does not give an extensive review of the field.Such reviews have appeared regularly; see, for example,