Short term forecasting of vegetables prices in Sicily

Virtually all of the Sicilian greenhouse vegetable production occurs in "cold" greenhouses and because of this quantities supplied vary significantly depending on the effects of climatic conditions on the production cycle. As a result price volatility tends to be extreme with significant variations from one week to the next and producers have to face unusually high risks and uncertainty associated with it. This is only compounded by the fact that vegetables have to be planted several months prior to marketing. It follows that good price forecasts could play a major role in assisting producers to make better planting and harvesting decisions. This paper presents the results of the application of time series forecasting models to Sicily's wholesale vegetable market prices. Specifically, univariate ARIMA (Autoregressive Integrated Moving Average) modelling techniques have been employed to produce price forecasts for six different vegetable products and the precision of the forecasts has been evaluated using standard criteria adopted in the literature. Possible developments to improve the performance of the models are also discussed following a critical examination of the results obtained.