Analyzing and Forecasting Weekly, Monthly and Quarterly Cotton Spot Prices: A Time Series Analysis Approach
MetadataShow full item record
Because of the increased variability of cotton spot over the past decade, the importance of accurate price forecasting for decision makes has increase. Three price series (weekly,monthly and quarterly) were moving average (ARIMA) models. The forecasting performance of ARIMA models was compared to that of other complex models such as random walk, simple exponential smoothing and moving average. Results showed that ARIMA models outperformed the random walk model and other simple forecasting techniques when forecasting monthly and quarterly prices. The weekly price series had an unstable stochastic structure. If constant re estimation of the weekly price model is not permitted the random walk model is a preferable forecasting method.