Forecast errors are shown for all forecasts Fig. 10 Sensitivity of forecasted life expectancy at birth of the LC-coherent model (right) and the proposed model (left) for Danish women (red) and men (blue) using the RC Sweden (solid lines) and, in addition, France, Italy, and Japan (dashed lines).
"We find that median forecasts of life expectancy are substantially larger than in existing official forecasts". The study is reported in the latest edition of the science journal Nature.
Our model forecasts increasing life expectancy at birth for both sexes; the observed values fluctuate (narrowly) around the median estimates.
Observed data are depicted in black Fig. 12 Prospective forecasts of life expectancy at birth for British (left) and Danish (right) men of the proposed model (blue), the P-spline approach (magenta), the LC model (red), and four of its variants: h0 (light red), h1 (yellow), h2 (green), and coherent (navy blue).
Nearly all forecasts had life expectancy too low, and so mortality too high, since the curves for the MAE and the ME are almost perfectly symmetric around zero.
Until relatively recently, he recalls, all the best brains in the world were forecasting that life expectancy would stall.
A forecast of life expectancy based on the average age at death of a person's four grandparents is a simple example of statistical forecasting.
Forecasts of active life expectancy: policy and fiscal implications.
Moreover, the forecasts of remaining life expectancy at age 65 are given for the proposed model in Figure 11 in the Appendix.
