Seasonal forecasting yield of wheat in India by feed-forward neural networks

Abstract

Wheat is the best principal food-grain of India and is the major food source of millions of Indians, awfully in the western, northern, north-eastern and north-western parts of the country. It has more in proteins, minerals, vitamins, fiber and carbohydrates and affords stable diet for human being. This paper demonstrations that a relative training is supported to explore the forecasting of yield per hectare of Wheat (in Kg) in India by means of Feed-Forward neural networks model and Box-Jenkins methods, which are surrounded by those forecasting models best magnificently practical ,functional and execute into practice. The yield of wheat data is collected from 1966 to 2012 for analysis of this model. The yield of wheat in India depends on two seasons like Kharif and Rabi. This training of yield of wheat examines the presentation of neural networks models and the results of which will be associated with those acquired by Box-Jenkins method. The Final conclusion of this paper is Feed-Forward neural networks models are better and superior than Box-Jenkins models.

Subject Classification:

• Statistics (62-07)
• Time series (37M10)
• Neural networks (92B20)

Keywords:

• Box-Jenkins methodology
• Wheat
• Food –grain
• Yield
• Feed –Forward Neural networks