Overview
John Maindonald is a statistician at the Statistical Consulting Unit, Graduate School, Australian National University. He is the author of Statistical Computation (Wiley, 1984) and contributed Poisson quasi-maximum-likelihood methodology — specifically the deviance criterion for adjusting k(t) to match observed death counts — to the Booth-Maindonald-Smith adaptation of Lee-Carter.
Key Contributions / Features
- Developed the deviance criterion (Maindonald 1984) used in Poisson quasi-ML estimation of k(t), adjusting the mortality index to match observed death counts D(x,t) each year rather than life expectancy alone.
- Co-authored the objective fitting-period selection framework (R(S)/RD(S) criterion) for Lee-Carter applied to Australian mortality data (1907–1999).
Related
Sources