Premium
Estimating the strength of expert judgement: The case of US mortality forecasts
Author(s) -
Alho Juha M.
Publication year - 1992
Publication title -
journal of forecasting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.543
H-Index - 59
eISSN - 1099-131X
pISSN - 0277-6693
DOI - 10.1002/for.3980110206
Subject(s) - judgement , estimation , process (computing) , computer science , econometrics , actuarial science , statistics , mathematics , economics , political science , management , law , operating system
The use of expert judgement is an important part of demographic forecasting. However, because judgement enters into the forecasting process in an informal way, it has been very difficult to assess its role relative to the analysis of past data. The use of targets in demographic forecasts permits us to embed the subjective forecasting process into a simple time‐series regression model, in which expert judgement is incorporated via mixed estimation. The strength of expert judgement is denned, and estimated using the official forecasts of cause‐specific mortality in the United States. We show that the weight given to judgement varies in an improbable manner by age. Overall, the weight given to judgement appears too high. An alternative approach to combining expert judgement and past data is suggested.