A Hybrid Grey Bayesian Rolling Model for Forecasting Annual Rural Unemployment Rate in West Bengal
209 / 143
Keywords:
Unemployment rate; Grey systems; Rolling grey model; Bayesian grey model; Forecasting; PLFS.Abstract
In this article, a reliable forecasting framework is developed for predicting the annual rural unemployment rate (UR) in West Bengal, India, where
time series data are scarce and noisy. Accurate unemployment forecasts are crucial for targeted policymaking, especially in rural regions facing
socio-economic vulnerabilities. The paper evaluates the performance of advanced grey system models in addressing data limitations and improving
predictive accuracy. The study utilizes annual rural UR estimates for various age groups and gender from 2017–18 to 2023–24, derived from the
Periodic Labour Force Survey (PLFS). Three forecasting models are developed: Grey Bayesian Model, Rolling Grey Model, and a hybrid Rolling
Grey Bayesian Model. The models are compared against traditional moving averages using Relative Mean Absolute Percentage Error (RMAPE).
Parameter estimation in the Bayesian models is implemented through MCMC simulations using WinBUGS. The results indicate that the Rolling Grey
and Rolling Grey Bayesian models consistently outperform traditional and standalone grey models, with RMAPE values as low as 8.40%. These
models show superior predictive capability, particularly for the 15–29 age group and female workforce categories. The dynamic updating mechanism
of the rolling models effectively handles short time series and chaotic labour market data. This study demonstrates the novel integration of grey
modelling with Bayesian inference and rolling mechanisms for rural labour market forecasting. It contributes a data-efficient, adaptive forecasting
approach suited for regions with limited labour statistics, offering practical value for policymakers in developing economies.
Downloads
References
1. Deng, J. (1989). Introduction to grey system theory. The Journal of
2. Grey System, 1(1), 1-24.
3. Akay, D. and Atak, M. (2007). Grey prediction with rolling mechanism
4. for electricity demand forecasting of Turkey. Energy, 32(9),
5. 1670-1675.
6. Arora, C. (2020). Impact of COVID-19 on Indian Unemployment:
7. A Study. Research Review Journals, 5(12), 31-36. https://doi.
8. org/10.31305/rrijm.2020.v05.i12.006
9. Basak, P., Ray, M., Sinha, K., and Anuja, A.R. (2023). Prediction of
10. Urban Unemployment Rate in India using Grey Model. Journal
11. of the Indian Society of Agricultural Statistics, 77(3), 243-248.
12. Ceylan, Z. (2021). Short-term prediction of COVID-19 spread using
13. grey rolling model optimized by particle swarm optimization.
14. Applied Soft Computing, 109, 107592. https://doi.org/10.1016/j.
15. asoc.2021.107592
16. Chakraborty, T., Chakraborty, A.K., Biswas, M., Banerjee, S., and
17. Bhattacharya, S. (2021). Unemployment rate forecasting: A
18. hybrid approach. Computational Economics, 57(1), 183-201.
19. Claveria, O. (2019). Forecasting the unemployment rate using the
20. degree of agreement in consumer unemployment expectations.
21. Journal for Labour Market Research, 53(1), 3. https://doi.
22. org/10.1186/s12651-019-0253-4
23. da Silva, L.A., and de Araújo, A.G. (2023). Effects of Unemployment
24. on Economic Sectors: A Proposal for Behavior Analysis with
25. Brazilian Municipalities. International Journal of Economics and
26. Finance, 15(9), 107.
27. Dahal, M.P., and Rai, H. (2019). Employment Intensity of Economic
28. Growth: Evidence from Nepal. Economic Journal of Development
29. Issues, 27(1), 34-47.
30. El-Fouly, T.H.M., El-Saadany, E.F., and Salama, M.M.A. (2007).
31. Improved grey predictor rolling models for wind power prediction.
32. IET Generation, Transmission & Distribution, 1(6), 928-937.
33. Hsu, L.C., and Wang, C.H. (2007). Forecasting the output of integrated
34. circuit industry using a grey model improved by the Bayesian
35. analysis. Technological Forecasting and Social Change, 74(6),
36. 843-853. https://doi.org/10.1016/j.techfore.2006.02.005
37. Li, B., Liu, F., Lin, J., and Wang, Z. (2020). Financial time series
38. forecasting model based on EMD and Rolling Grey Model. In
39. 2020 IEEE Workshop on Signal Processing Systems (SiPS) (pp.
40. 1-6). IEEE.
41. Mimi, M.B., Haque, M.A.U., and Kibria, M.G. (2022). Does human
42. capital investment influence unemployment rate in Bangladesh:
43. A fresh analysis. National Accounting Review, 4(3), 273-286.
44. https://doi.org/10.3934/NAR.2022016
45. Nair, S. (2020). A study on the causes and impact of unemployment in
46. India. International Review of Business and Economics, 4(2), 53.
47. Nguyen, P.H., Tsai, J.F., Kayral, I.E., and Lin, M.H. (2021).
48. Unemployment rates forecasting with grey-based models in the
49. post-COVID-19 period: A case study from Vietnam. Sustainability,
50. 13(14), 7879. https://doi.org/10.3390/su13147879
51. Pavithra, V., Gupta, D.S., Basak, P., Debnath, M.K., and Mula, G.
52. (2024). Prediction of Annual Rural Unemployment Rate in
53. West Bengal using Grey Model. Journal of the Indian Society
54. of Agricultural Statistics, 78(1), 47-52. https://doi.org/10.56093/
55. JISAS.V78I1.6
56. Sinha, K., and Sahu, P.K. (2020). Forecasting Short Time Series using
57. Rolling Grey Bayesian Framework. International Journal of
58. Statistical Sciences, 20(2), 207-224.
59. Syafwan, H., Putri, P., and Syafwan, M. (2023). Forecasting
60. Unemployment in Indonesia using Weighted Moving Average
61. Method. JURTEKSI (Jurnal Teknologi dan Sistem Informasi),
62. 9(4), 699-706. https://doi.org/10.33330/jurteksi.v9i4.2624