Time Series Forecasting of Suicide Mortality in Europe: A Comparative Performance Analysis of ARIMA, Holt-Winters Exponential Smoothing, and Naive Models (1950--2019)
DOI:
https://doi.org/10.54536/ajmsi.v5i1.6363Keywords:
ARIMA Model, Holt--Winters Exponential Smoothing, Naive Model, Public Health Surveillance, Suicide Mortality, Time Series ForecastingAbstract
This paper evaluates the forecasting performance of three common time series approaches: ARIMA, Holt--Winters exponential smoothing, and the naive model using annual suicide mortality data from Europe covering 1950 to 2019 (Box & Jenkins, 2015). The European region offers unusually long and consistent mortality records, which creates a valuable opportunity to examine how different forecasting methods behave when data quality and continuity are relatively strong. The dataset (70 observations) was divided into a training period (1950--2010) and a test period (2011--2019). The models were compared using several accuracy indicators, including MAE, RMSE, MAPE, and MASE. Among the three approaches, ARIMA(1,1,3) provided the most accurate forecasts, with considerably lower errors compared to Holt--Winters and the naive benchmark (Box et al., 2015). The strong performance of ARIMA appears closely linked to the stable autocorrelation and the long historical span of the European series, which favor models capable of capturing both lagged dependencies and persistent changes over time. The findings suggest that model choice should depend on the specific nature of the data rather than relying on general preferences or assumptions.
Downloads
References
Box, G. E. P., & Jenkins, G. M. (1976). Time series analysis: Forecasting and control (Rev. ed.). Holden-Day.
Box, G. E. P., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: Forecasting and control (5th ed.). Wiley.
Chatfield, C., Koehler, A. B., Ord, J. K., & Snyder, R. D. (2001). A new look at models for exponential smoothing. Journal of the Royal Statistical Society: Series D, 50(2), 147--159. https://doi.org/10.1111/1467-9884.00264
Gunnell, D., Cowan, S., O’Connor, R. C., & Kapur, N. (2023). The global financial crisis and suicide: Systematic review of correlations and causal mechanisms. BJPsych Open, 8(2), e47.
Hyndman, R. J. (2014). Measuring forecast accuracy. In M. Gilliland, L. Tashman, & U. Sglavo (Eds.), Business forecasting: Practical problems and solutions (pp. 177--183). Wiley.
Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: Principles and practice (3rd ed.). OTexts.
Hyndman, R. J., & Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4), 679--688. https://doi.org/10.1016/j.ijforecast.2006.03.001
Hyndman, R. J., & Rostami-Tabar, B. (2024). Forecasting interrupted time series. International Journal of Forecasting, 40(2), 675--690. https://doi.org/10.1016/j.ijforecast.2023.06.003
Hyndman, R. J., Koehler, A. B., Ord, J. K., & Snyder, R. D. (2008). Forecasting with exponential smoothing: The state space approach. Springer. https://doi.org/10.1007/978-3-540-71918-2
McNown, R., & Rogers, A. (1992). Forecasting cause-specific mortality using time series methods. International Journal of Forecasting, 8(3), 413--432. https://doi.org/10.1016/0169-2070(92)90055-E
Pleños, M., Gómez, J., & Santos, R. (2022). Time series forecasting using Holt--Winters exponential smoothing: Applications to environmental data. Procedia Computer Science, 207, 2841--2850. https://doi.org/10.1016/j.procs.2022.09.334
Riaz, M., Rauf, A., & Khan, M. S. (2023). Epidemiological forecasting models using ARIMA, SARIMA, and Holt--Winters methods: Application to malaria cases. Mathematical Biosciences and Engineering, 20(9), 16857--16878. https://doi.org/10.3934/mbe.2023790
Turecki, G., Brent, D. A., Gunnell, D., O’Brien, P. M., & Oquendo, M. A. (2019). Suicide and suicidal behaviour. The Lancet, 387(10024), 1227--1239.
Winters, P. R. (1960). Forecasting sales by exponentially weighted moving averages. Management Science, 6(3), 324--342. https://doi.org/10.1287/mnsc.6.3.324
World Health Organization. (2023). Mortality data. Retrieved from https://www.who.int/data/mortality
Yang, W., Li, J., & Chen, Y. (2023). Application of exponential smoothing method and SARIMA model in forecasting hospital admissions. BMC Medical Informatics and Decision Making, 23, 256. https://doi.org/10.1186/s12911-023-02156-7
Zeng, Q., Li, D., & Zhang, Z. (2014). Forecasting hospital bed demand using ARIMA and double exponential smoothing models. Mathematical Problems in Engineering, 2014, 1--9. https://doi.org/10.1155/2014/214058
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Mehrez Ben Nasr, Sirine Ben Othman
This work is licensed under a Creative Commons Attribution 4.0 International License.
