Regression Model Selection and Assessment of Agricultural Water Price Affordability in China

Abstract

The agricultural water price depends on the agricultural water price affordability (AWPA) in each region. This study found that the logarithmic linear model had the best fitting effect through evaluating the grey situation decision model, which considered factors such as rainfall and output value per unit area. The contribution of each influencing factor was determined by the Lindeman–Merenda–Gold method. We established a new model to determine the water expenditure coefficient (WEC) by improving the way that the value of the WEC is assigned. Then, the AWPA in different regions was calculated. The results showed that the WEC was between 2.62% and 12.95%, and the AWPA price was between 0.058 and 0.52 yuan/m3 (0.0084 and 0.075 $/m3). The contribution of precipitation and output was 45.20% and 25.60%, respectively. The WEC and AWPA in northeast, northwest, and northern China are higher than those in southwest and southern China. The AWPA in the Yellow River Basin was higher than that in the Yangtze River Basin; however, the space for adjustment in the Yellow River Basin was slightly smaller than that in the Yangtze River Basin.