Inverse Projection
Also known as: Back projection, Generalized inverse projection, Demographic back-projection, Lee-Wrigley-Schofield projection
Inverse projection, and its more flexible successor generalized inverse projection, reconstructs the demographic history of a population from the outside in. Where conventional cohort-component projection runs a known population forward using assumed rates, inverse projection runs the logic backward: starting from a population of known size and age structure at one date, and given annual totals of births and deaths, it infers the population sizes, age distributions, mortality levels, life expectancies and fertility rates that must have prevailed in earlier years. The technique was devised by Ronald Lee and applied by Wrigley and Schofield to their English aggregative series, allowing three centuries of population history to be recovered without any direct census before 1801. Jim Oeppen's generalization relaxed restrictive assumptions about migration and closed populations. The method is the bridge that turns raw counts of vital events into a fully articulated demographic regime.
Key highlights
- Recovers full demographic regimes from aggregate flows without regular censuses
- Internally consistent: outputs satisfy birth, death and population accounting identities
- Generalized versions accommodate net migration and two-sided boundary conditions
- Turns inexpensive aggregative series into rich age-structured reconstructions
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use inverse projection when you possess long, reliable annual series of total births and deaths and at least one trustworthy population benchmark, but lack the regular censuses needed for direct demographic measurement. It is the standard tool for converting aggregative parish-register series into full demographic reconstructions—population size, age structure, life expectancy and fertility—across the pre-census era. It suits national or large regional populations where migration is modest or can be estimated, and where a model life-table family plausibly captures the mortality pattern. It is inappropriate for small, highly migratory populations, or where the birth and death series are too short, too incomplete, or too poorly dated to support reverse cohort accounting.
Strengths & limitations
- Recovers full demographic regimes from aggregate flows without regular censuses
- Internally consistent: outputs satisfy birth, death and population accounting identities
- Generalized versions accommodate net migration and two-sided boundary conditions
- Turns inexpensive aggregative series into rich age-structured reconstructions
- Depends on the assumed model life-table family, which imposes a mortality shape
- Sensitive to errors and under-registration in the input birth and death series
- Classic back projection assumes a closed population, mishandling migration
- Requires an accurate anchoring population; benchmark error propagates throughout
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How does it differ from ordinary projection?
Standard cohort-component projection takes a known population and assumed future rates and computes the population forward. Inverse projection reverses the unknown: it takes observed aggregate births and deaths plus an anchoring population and infers the past populations, age structures and rates consistent with them. One predicts the future from rates; the other reconstructs the past from flows and a benchmark.
Why are model life tables necessary?
A single annual total of deaths cannot reveal how mortality was distributed across ages. Model life tables supply that age pattern from a family of empirically derived schedules indexed by one level parameter. The method then solves for the level each year that reproduces the observed deaths. Without this borrowed shape, converting aggregate deaths into age-specific survival would be underdetermined.
What did generalized inverse projection fix?
Classic back projection assumed a closed population and worked only from a terminal census backward, mishandling migration and depending on one boundary. Oeppen's generalized version lets net migration enter the accounting and permits boundary conditions at both ends of the series, improving stability and realism. It made the previously implicit closure assumptions explicit and allowed reconstructions for populations with significant migration.
Sources
- 1.Wrigley, E. A., & Schofield, R. S. (1981). The Population History of England 1541-1871: A Reconstruction. Edward Arnold / Harvard University Press.ISBN 9780674690073
- 2.Wrigley, E. A., Davies, R. S., Oeppen, J. E., & Schofield, R. S. (1997). English Population History from Family Reconstitution 1580-1837. Cambridge University Press.ISBN 9780521590150
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Inverse Projection. ScholarGate. https://scholargate.app/historical-demography/inverse-projection-demography