Extending the Heston model to forecast motor vehicle collision rates

We present an alternative approach to the forecasting of motor vehicle collision rates. adopt oft-used tool in mathematical finance, Heston Stochastic Volatility model, forecast short-term and long-term evolution incorporate a number extensions model make it fit for modelling temporally-unstable non-deterministic nature rate fluctuations, introduce parameter account periods accelerated safety. also adjust estimates seasonality patterns. Using these parameters, we perform rates explore plausible scenarios using forecasts. The shows close affinity with realised (over 95% accuracy), outperforms models currently used road safety research (Vasicek, SARIMA, SARIMA-GARCH). suggest that modest targets reduce (1.83% annually) fluctuations month-to-month (by half) could have significant benefits median this scenario suggests 50% fall rates, 75% simulations suggesting effective change is observed before 2044. main benefit provides eschewing necessity setting unreasonable are often missed. Instead, presents effects achievable can on over long run, while incorporating random variability. Examining parameters underlie expected will aid policymakers determining effectiveness implemented policies.