Actuarial-Informed Neural Networks: Constrained Deep Learning for Sex-Specific Multi-Population Longevity Forecasting
Organizations: Centre of Excellence for Global Finance Operations and Transformation Swiss Re Management Ltd∗ Mythenquai 50/60, 8022 Zurich, Switzerland
Abstract
We introduce Actuarial-Informed Neural Networks (AINN), a constrained deep learning framework for sex-specific multi-population longevity forecasting that embeds actuarial knowledge into LSTM training through differentiable penalty terms. Applied to a high-longevity 6-country cluster (Switzerland, Sweden, Norway, West Germany, Netherlands, Japan; 1956-2020), the framework is a challenger to the Li-Lee benchmark and unconstrained neural approaches. Our central finding is that monotonicity constraint intensity lambda functions as a continuous governance dial with a quantifiable accuracy-stability trade-off. At lambda = 0.01, 30-year forecast confidence intervals narrow by 6.5% at zero RMSE cost. At lambda = 0.1, they narrow by 25% at only +0.84% RMSE cost. This trade-off is invisible to standard hyperparameter optimizers, which select lambda by minimizing one-step RMSE, the wrong metric for structural constraints in recursive forecasting. The framework adds three contributions: (i) residual-calibrated process noise that eliminates double-counting in the dual uncertainty framework, reducing confidence intervals by 53-60%; (ii) a 5-seed model ensemble with a coefficient of variation of 1.06%; and (iii) a time-varying credibility weight Z(t) = exp(-alpha t) connecting to Buhlmann credibility theory for blending neural and actuarial predictions. A 10-year out-of-sample backtest (trained on 1956-2010, validated against observed HMD life expectancy for 2011-2020) yields a mean absolute error of 0.38 years, with 11 of 12 country x sex combinations within the 90% confidence interval. The model produces regulatory-grade outputs including Solvency Capital Requirements under the Swiss Solvency Test (ES 99.0%: +1.91 years, CHE Male) and Solvency II, reverse stress testing, and model-based stability verification (amplification ratio 1.02x).
Figures & tables
| Country | Combined [ 9 ] | Male | Female |
|---|---|---|---|
| Switzerland | Conflict | Stationary | Conflict |
| Sweden | Unit Root | Stationary | Stationary |
| Norway | Stationary | Conflict | Conflict |
| West Germany | Unit Root | Unit Root | Unit Root |
| Netherlands | Unit Root | Unit Root | Unit Root |
| Japan | Conflict | Conflict | Conflict |
| Parameter | Values | Rationale |
|---|---|---|
| Lookback | {5, 7, 10, 12, 15} | Temporal context |
| Units | {16, 32, 48, 64} | First LSTM capacity |
| Units | {8, 16, 24, 32} | Second LSTM capacity |
| Learning rate | {0.01, 0.005, 0.001, 0.0005, 0.0001} | Optimizer step |
| {0, 0.001, 0.01, 0.1, 1.0} | Coherence weight | |
| {0, 0.001, 0.01, 0.1, 1.0} | Monotonicity weight |
| Parameter | Value |
|---|---|
| Lookback | 15 years |
| LSTM layers | 48, 32 units |
| Learning rate | 0.001 |
| 0.001 | |
| 0.001 | |
| Validation RMSE | 6.1725 |
| Factor | Reduction | ||
|---|---|---|---|
| (Male) | 5.99 | 2.80 | |
| (Female) | 9.08 | 3.64 |
| Configuration | RMSE |
|---|---|
| Separate M/F, fixed 32-16, lb=10 | 7.77 |
| + Joint M/F ( ) | 7.59 |
| + Architecture tuning via Keras Tuner ( ) | 7.07 |
| + Lookback lb=15 via Optuna ( ) | 6.17 |
| + Constraints ( ) | 6.17 |
| RMSE (cost) | frac( ) | CI M | CI F | CI Reduction | |
|---|---|---|---|---|---|
| 0.0 | 6.173 (—-) | 55.6% | 3.64 | 3.11 | — |
| 0.001 | 6.173 ( 0.01%) | 55.6% | 3.62 | 3.10 | 0.4% |
| 0.01 | 6.171 ( 0.03%) | 55.6% | 3.40 | 3.03 | 6.5% |
| 0.1 | 6.225 ( 0.84%) | 44.4% | 2.70 | 2.43 | 25.7% |
| 1.0 | 6.315 ( 2.29%) | 27.8% | 1.93 | 1.71 | 46.8% |
| Country | Sex | (2020) | (2050) | CI Width | Gain |
|---|---|---|---|---|---|
| Switzerland | M | 80.26 | 82.06 | 3.69 | +1.80 |
| Switzerland | F | 83.58 | 85.55 | 3.03 | +1.96 |
| Sweden | M | 79.94 | 81.74 | 3.73 | +1.80 |
| Sweden | F | 82.93 | 85.01 | 3.26 | +2.08 |
| Norway | M | 80.58 | 82.30 | 3.57 | +1.73 |
| Norway | F | 83.27 | 85.28 | 3.13 | +2.01 |
| Sex | Model | (2050) | Gain | CI Width | SCR (ES) |
|---|---|---|---|---|---|
| M | Li-Lee | 83.54 | +3.28 | 3.73 | +2.091 |
| M | AINN | 82.06 | +1.80 | 3.69 | +1.907 |
| F | Li-Lee | 86.05 | +2.47 | 4.71 | +2.178 |
| F | AINN | 85.55 | +1.96 | 3.03 | +1.516 |
| Country | Sex | Observed | AINN | Li-Lee | AINN Error | In 90% CI |
|---|---|---|---|---|---|---|
| CHE | M | 80.26 | 80.52 | 80.51 | 0.26 | YES |
| CHE | F | 83.58 | 83.84 | 84.04 | 0.26 | YES |
| NOR | M | 80.58 | 79.56 | 79.54 | 1.02 | NO |
| JPN | F | 84.89 | 84.78 | 84.95 | 0.11 | YES |
| Country | Male SCR | Female SCR |
|---|---|---|
| Switzerland | +1.907 | +1.516 |
| Sweden | +1.942 | +1.649 |
| Norway | +1.852 | +1.580 |
| W. Germany | +2.245 | +1.800 |
| Netherlands | +2.063 | +1.849 |
| Japan | +1.902 | +1.243 |
| (2025) | (2035) | (2050) | M Med | F Med | M CI | |
|---|---|---|---|---|---|---|
| 0.0 | 1.000 | 1.000 | 1.000 | 82.06 | 85.55 | 3.69 |
| 0.02 | 0.905 | 0.741 | 0.549 | 81.69 | 85.14 | 2.89 |
| 0.05 | 0.779 | 0.472 | 0.223 | 81.26 | 84.73 | 2.16 |
| 0.10 | 0.607 | 0.223 | 0.050 | 80.90 | 84.33 | 1.81 |
| 0.20 | 0.368 | 0.050 | 0.002 | 80.63 | 83.98 | 1.35 |
Mortality Prediction
Mortality prediction estimates the risk, timing, or cause of death from clinical records, images, signals, and population data.
See all 33 papers