Forecasting crop water demand
MetadataShow full item record
The dearth of information and invalid modeling approaches restrict crop water demand forecasting and thereby, efficient allocation of limited water resources in Georgia. This study adopts econometrics, structural time series (STMS), and univariate time series (ARIMA) approaches to forecast corn and soybeans irrigation water demand. The expected utility maximization theory and crop acreage supply response models offer the theoretical backup needed for the study. Especial efforts have been made to improve the existing crop acreage response models by incorporating institutional variables and stochastic trend variable. In our analysis, economic and institutional variables yield expected signs and significant results reflecting the importance of these variables in crop water demand forecasting. Moreover, inclusion of different nature of trend variable also improves the corn and soybeans acreage supply models. Further analysis using ARIMA and STSM with stochastic trend and no explanatory variables presents pure statistical perspective of the water demand forecasting issue.