Knowledge Can Improve Forecasts: A Review of Selected Socioeconomic Population Projection Models

Abstract

Since Nathan Keyfitz wrote that knowledge could not improve forecasts there has been a presumption that causal modeling has little to contribute to demographic forecasting. However this presumption should be reversed since combining forecasts from causal models with those of the UN produced more accurate predictions than those of the UN. This occurs because the forecast errors from the causal models and those of the UN are not highly correlated and because the forecasts from the causal model are not less accurate than those of the UN. To support this statement three different types of causal models were considered: World3 model Bachue-Philippines and Bachue-Kenya models and the Wheeler model. The three sections in this paper demonstrate how the application of causal models can improve demographic projections.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 43 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.