Abstract:
The farmers in some jackfruit growing countries are getting a good return from jackfruit than
other major fruits. By means of bargaining, the price is fixed depending upon the supply,
demand, quality of jackfruit, etc. (APAARI, 2012). In most cases farmers are uncertain about
their future income due to fluctuations in price and / or government policies. Predicting the
price of jack fruit and identification of seasonal month/s supports farmers to earn income and
it is a key element in their decision makings. Therefore, this study focused to fix an
appropriate model on price forecasting of Jack fruit in future. The study area was Jaffna
district where Jack fruit is highly obtained from the home gardens. Wholesale prices of Jack
fruit were accumulated from year 2008-2014 from Department of Agriculture, Jaffna. Time
series plot of prices was observed initially and ensured that data are not stationary.
Therefore, Seasonal Auto-Regressive Integrated Moving Average (SARIMA) model was
applied. The P-value of both of estimated 'parameters and Box-Pierce chi-squared statistics
were hypothesized. Appropriate price forecasting model was fixed after comparisons of error
measurements among various SARIMA models. The model which had the lowest value of
error terms was selected as the best model for forecasting of prices. Finally, time series plot
was observed between actual value and forecasted value after model fitting. Also,
multiplicative decomposition analysis was performed to observe seasonal month/s. Major
findings revealed that price of jack fruit had been increasing from year 2008-2014 with
seasonal fluctuations. SARIMA(1, 1, 1) (1, 0, 0) is the best model for forecasting of prices of 12
jack fruit. Jack fruit is vastly obtainable during the period of April-July. The price forecasting
of jack fruit helps to farmers to achieve high sales of return while evading uncertainty of
income.
