Forecasting Labour Force Trends Among Older Persons in Malaysia using Time Series Analysis

Abstrak

Malaysia has transitioned into an ageing society faster than the previous demographic model suggested. The latest report from the Department of Statistics Malaysia (DOSM) confirms that as of 2025, individuals aged 65 and older already comprise 8% of the total population. To address the objectives of Sustainable Development Goal (SDG) 8.5 regarding productive employment, this study forecasts labour force participation trends among Malaysians aged 60 to 64 through the year 2030. This specific age group represents a critical segment for extending working lives and maintaining national productivity. The analysis utilises annual time series data from 1982 to 2021 for model estimation and evaluation, while actual observations from 2022 to 2024 serve as an ex-post benchmark to verify the forecast accuracy. This study applies four forecasting techniques, including double exponential smoothing (DES), Holt’s exponential smoothing (HES), autoregressive integrated moving average (ARIMA), and time series regression (TSR). Following evaluation via mean absolute percentage error (MAPE), root mean square error (RMSE), and geometric root mean square error (GRMSE), the HES emerged as the most reliable, achieving a precision rate of 99.27% (0.73% error) against 2024 actuals. The final forecast trends indicate steady expansion, with the labour force participation expected to reach 533,020 older workers by 2030, which is a 10.47% cumulative increase from 2024. These findings confirm that prolonged workforce participation is no longer a temporary shift but a structural reality. Consequently, Malaysia requires immediate policy interventions focused on flexible retirement frameworks, targeted reskilling, and the creation of an age-inclusive workplace environment to sustain economic stability.