2023, Vol. 8, Special Issue 4, Part H
Forecasting agricultural commodity prices using singular spectrum analysisAuthor(s):
Prabhat Kumar, Girish Kumar Jha, Rajeev Ranjan Kumar and Achal LamaAbstract:
In recent times, the effectiveness of appropriate time series decomposition techniques has gained large acceptance. Among various time series decomposition techniques, singular spectrum analysis (SSA) is a highly promising technique. Its successful application has been demonstrated across various contexts, showcasing its efficacy in effectively separating and understanding different components within time series data. Thus, in this study, we have used SSA and its forecasting method SSA-LRF for modelling and forecasting agricultural price series, namely tomato of two markets, i.e., Delhi and Lucknow. Further, the results obtained from SSA-LRF are compared with that of the SSA-ARIMA and ARIMA models. The comparative analysis was carried out using RMSE, MAPE, and MAE criteria. We report the superiority of the SSA-LRF model over others under consideration in terms of the lowest RMSE, MAPE, and MAE values. This study has highlighted the importance of decomposition-based forecasting techniques such as SSA-LRF for agriculture price series. Pages: 586-591 | Views: 85 | Downloads: 5Download Full Article: Click Here
How to cite this article:
Prabhat Kumar, Girish Kumar Jha, Rajeev Ranjan Kumar, Achal Lama. Forecasting agricultural commodity prices using singular spectrum analysis. Int J Stat Appl Math 2023;8(4S):586-591.