Fruit Trade in India. Matlab Programming for Prediction of monthly Arrival and Prices of various Fruits

Project Report from the year 2009 in the subject Engineering - General, Basics, grade: Bachelor of Technology, , language: English, abstract: In this study an attempt was made to analyse the fruit arrival and price patterns in a wholesale fruit market (Haldwani Mandi) for trend and seasonality in order to develop forecasting models for the fruit arrival process so as to rationalize an important input to fruit mandi system design. Historical time series data on monthly arrivals and average monthly prices was collected from the Haldwani mandi records for the period January, 1990 to April, 2009. Arrivals and prices of five fruits (Mango, Banana, Apple, Peach and Orange) were considered. A program in MATLAB 7.0 was developed for trend and seasonal analysis. For forecasting, these trend models were extended and seasonal index were applied for each month. Forecasting models were developed on the basis of first 204 months (January 1990 to December 2006) data using time series analysis technique. Forecasts were generated for the next 36 months (January 2007 to December 2009). These forecasts were compared with the actual arrivals for January 2007 to April 2009. Results of the present investigation have shown that arrival of Apple was observed in the month of July, Banana arrived maximum in October, whereas Mango, Orange and Peach recorded maximum arrival in the months of July, April and May respectively. Seasonal indices corresponding to these peak arrivals were 626.66, 136.29, 876.06, 371.17 and 439.89 respectively. In the case of Apple, arrivals and prices were inversely correlated; i.e., prices were low during peak arrival. However, in the case of banana, there was very little correlation; one major reason for this seems to be that the prices of Banana did not vary throughout the year. There was a increasing trend in the case of prices. In the trend analysis, it was observed that the arrival of Apple and Orange had a decreasing trend for the years under analysis, wher