Authors : Karuppasamy K. M; K. Senthamarai Kannan

Volume/Issue : Volume 8 - 2023, Issue 9 - September

Google Scholar : https://bit.ly/3TmGbDi

Scribd : https://tinyurl.com/ytxb7fkh

DOI : https://doi.org/10.5281/zenodo.8376502

Abstract : Time series analysis is one of the utmost popular forecasting techniques in predicting the upcoming events based on preceding performance. In this paper, monthly regular prices of Rice data were used in seasonal auto regressive integrated moving average (SARIMA), simple exponential smoothing model and naïve method incorporated to predict the upcoming prices of Rice and thereby compared. SARIMA and simple exponential smoothing model was found appropriate for all Indian rice price. The performance evaluation of the fitted model was observed by calculating various measures. The SARIMA model was the most typical model at the cost estimation of paddy in by and large India. Those models were best fitted for predicting of Rice Price in India. Prediction was made for the immediate next two years that is 01-06-2021 to 01-05- 2023. The performance evaluation of these models were validated by comparison with percentage deviation from the actual values and mean absolute error (MAPE), which was found to be 4.18% for the price under rice in India.

Keywords : SARIMA, Simple Exponential Smoothing, Naïve Model, AIC, BIC and Forecasting.