EUVIMEDEUVIMED entdecken
Crossref

A Neural Network-Driven Hybrid Time Series Model for Yield Forecasting of Finger Millet in Odisha

Subrat Kumar Mahapatra, Digvijay Singh Dhakre, Debasis Bhattacharya, Chapala Bohidar

INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES · 2026-06 · 22 (01) · 173

DOI: 10.59467/ijass.2026.22.173 ↗

Kurz gefasst

Fachartikel in der Zeitschrift INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES, 2026-06. Verfasst von Subrat Kumar Mahapatra und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The present investigation has been carried out for the yield forecasting of finger millet by using the Neural Network based Hybrid time series forecasting model. In this study, different statistical and machine learning models like Autoregressive Integrated Moving Average (ARIMA), Neural Network Autoregression (NNAR), Support Vector Machines (SVM) and Hybrid Models are used for crop yield predictions. Various models have been identified based on the different parameters such as the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The study reveals that the Hybrid ARIMA (1,1,1) + NNAR (2,2,) is found to be the best fitted model due to the lowest value of RMSE and MAPE for the training data and also for the testing data, among all the comparing model. The best fitted Hybrid model is used to forecast the yield of millet for next 8 years in Odisha. From this study, it has been found that the yield of millet is expected to follow a cyclic trend in coming years, in few years it is expected to increase, and in few years, it is expected to decrease. This study will help the policy makers, government in decision making and also plays an important role in determining the yield gap, production and demand gap and differences for different millet in Odisha in the coming years.. KEYWORDS :Neural network, Hybrid model, Forecasting, Yield, Millet.

Zitieren

APA

Mahapatra, S. K., Dhakre, D. S., Bhattacharya, D., & Bohidar, C. (2026). A Neural Network-Driven Hybrid Time Series Model for Yield Forecasting of Finger Millet in Odisha. INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES, 22(01), 173. https://doi.org/10.59467/ijass.2026.22.173

Vancouver

Mahapatra SK, Dhakre DS, Bhattacharya D, Bohidar C. A Neural Network-Driven Hybrid Time Series Model for Yield Forecasting of Finger Millet in Odisha. INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES. 2026;22(01):173. doi:10.59467/ijass.2026.22.173

BibTeX
@article{Mahapatra2026,
  author = {Mahapatra, Subrat Kumar and Dhakre, Digvijay Singh and Bhattacharya, Debasis and Bohidar, Chapala},
  title = {A Neural Network-Driven Hybrid Time Series Model for Yield Forecasting of Finger Millet in Odisha},
  journal = {INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES},
  year = {2026},
  volume = {22},
  number = {01},
  pages = {173},
  doi = {10.59467/ijass.2026.22.173}
}
Kostenlos starten →