Hybrid ARFIMA-LRNN Model for Forecasting Commodity Prices

Debopam Rakshit; Ranjit Kumar Paul

doi:10.56093/jisas.v78i03.171405

Authors

Debopam Rakshit ICAR-Indian Veterinary Research Institute, Izatnagar
Ranjit Kumar Paul ICAR- Indian Agricultural Statistics Research Institute, New Delhi

https://doi.org/10.56093/jisas.v78i03.171405

Keywords:

ARFIMA; Hybrid model; Long memory; Neural network; RNN.

Abstract

The unexpected fluctuation of prices of agricultural commodities may have impactful repercussions on the producers. The price volatility can be modeled by applying time series analysis. A time series consists of linear and non-linear components. The linear component can be modeled by the autoregressive integrated moving average (ARIMA) methodology. Again, long-term dependencies amongst the realizations of any time series can
also be observed. This long-term dependency can be addressed by incorporating fractional differencing in the ARIMA model which is known as the autoregressive fractionally integrated moving average (ARFIMA) model. To address both the linear and non-linear components effectively, hybrid time series models can be used. In a hybrid model, more than one model is clubbed together such that one is used for capturing the linear component and another captures the nonlinear counterpart. In this article, a hybrid ARFIMA-LRNN (Layer recurrent neural network) model is employed for modeling the price series of arhar for the Mumbai market. The forecasting accuracy of the hybrid model outperformed the standalone models.

Downloads

Download data is not yet available.

References

Box, G.E., Jenkins, G.M., Reinsel, G.C., and Ljung, G.M. (2015). Time series analysis: forecasting and control. John Wiley & Sons, New Jersey, USA.

Brockwell, P.J., and Davis, R.A. (1991). Time Series: Theory and Methods (2nd ed.). Springer, New York.

Broock, W.A., Dechert, W.D., and Scheinkman, J.A. (1996). A test for independence based on the correlation dimension. Econometric Reviews, 15(3), 197-235.

Dickey, D., and Fuller, W. (1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of American Statistical Association, 74, 427-431.

Elman, J.L. (1990). Finding structure in time. Cognitive Science, 14, 179-211.

Garai, S., Paul, R.K., Rakshit, D., Yeasin, M., Emam, W., Tashkandy, Y., and Chesneau, C. (2023). Wavelets in combination with stochastic and machine learning models to predict agricultural prices. Mathematics, 11(13), 2896.

Garai, S., Paul, R.K., and Paul, A.K. (2024). Spillover effects of Covid-19 induced lockdown on onion prices in India. Journal of Scientific Research and Reports, 30(3), 21-31.

Geweke, J., and Porter-Hudak, S. (1983). The estimation and application of long memory time series models. Journal of Time Series Analysis, 4(4), 221-238.

Granger, C.W.J., and Joyeux, R. (1980). An introduction to long memory time series models and fractional differencing. Journal of Time Series Analysis, 1(1), 15-29.

Hosking, J.R.M. (1981). Fractional differencing. Biometrika, 68, 165-176.

Levenberg, K. (1944). A method for the solution of certain non-linear problems in least squares. Quarterly of Applied Mathematics, 2(2), 164-168.

Phillips, P.C.B., and Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335-346.

Liu, Q., Guo, Z., and Wang, J. (2012). A one-layer recurrent neural network for constrained pseudoconvex optimization and its application for dynamic portfolio optimization. Neural Networks, 26, 99-109.

Marquardt, D. (1963). An algorithm for least-squares estimation of nonlinear parameters. SIAM Journal on Applied Mathematics, 11(2), 431-441.

Mitra, D., Paul, R.K., Paul, A.K., and Bhar, L.M. (2018). Forecasting time series allowing for long memory and structural break. Journal of the Indian Society of Agricultural Statistics, 72(1), 49-60.

Paul, R.K. (2014). Forecasting wholesale price of pigeon pea using long memory time-series models. Agricultural Economics Research Review, 27(2), 167-176.

Paul, R.K., Gurung, B., and Paul, A.K. (2015). Modelling and forecasting of retail price of arhar dal in Karnal, Haryana. The Indian Journal of Agricultural Sciences, 85(1), 69-72.

Paul, R.K., Yeasin, M., Kumar, P., Kumar, P., Balasubramanian, M., Roy, H.S., ... and Gupta, A. (2022). Machine learning techniques for forecasting agricultural prices: A case of brinjal in Odisha, India. PLOS ONE, 17(7), e0270553.

Paul, R.K., Yeasin, M., Kumar, P., Paul, A.K., and Roy, H.S. (2023). Deep learning technique for forecasting the price of cauliflower. Current Science, 124(9), 1065-1073.

Pwasong, A., and Sathasivam, S. (2018). Forecasting comparisons using a hybrid ARFIMA and LRNN models. Communications in Statistics – Simulation and Computation, 47(8), 2286-2303.

Rakshit, D., Paul, R.K., and Panwar, S. (2021). Asymmetric price volatility of onion in India. Indian Journal of Agricultural Economics, 76(2), 245-260.

Rakshit, D., and Paul, R.K. (2024a). Development of out-of-sample forecast formulae for the FIGARCH model. Model Assisted Statistics and Applications, 19(2), 133-143.

Rakshit, D., and Paul, R.K. (2024b). Modeling time series with asymmetric volatility and long memory. Indian Journal of Agricultural Economics, 79(2), 231-244.

Shapiro, S.S., and Wilk, M.B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3/4), 591-611.

Tamilselvi, C., Yeasin, M., Paul, R.K., and Paul, A.K. (2024). Can denoising enhance prediction accuracy of learning models? A case of wavelet decomposition approach. Forecasting, 6(1), 81-99.

Tsai, H. (2006). Quasi-maximum likelihood estimation of long memory limiting aggregate processes. Statistica Sinica, 16(1), 213-226.