The main objective of this study is to analyze the demand for water in Denton. The data used for the study are obtained from the City of Denton Utilities Department, the Tax Appraisal District and government documents. The 121 households which have perfect ten years historical data of water consumption were selected to be the representatives of all households in Denton. The study reveals that the change in water consumption significantly relates to the change in marginal price. Furthermore, the weather variables also have strong effects on the water consumption, especially during summer. The coefficients of income and a "difference" variable are found to have the opposite sign but are not equal in magnitude. In fact, they should be equal in magnitude, but opposite in sign. While the estimated coefficients on all independent variables were highly significant statistically, the resulting coefficient on the house size variable was statistically insignificant in the model test. The results show that the difference variable is required in the model. It also had some effect on the water consumption. It is found that there is a small change in water consumption when the lot size is increased.

This paper analyzes the demand for water in case of a multiple-part tariff in Denton, Texas. The model used is developed from Billing & Agthe's model by using the following variables: marginal price, difference variable, tax assessed value, lot size, house size, temperature and rainfall.. The results indicate that temperature has the greatest effect on water demand, since this area is considered to be a very warm area. Also, marginal price seems to have a strong effect on water consumption indicating that customer is well-informed to a change in rate schedule. This test supports the original idea of the previous articles that the coefficient on difference variable and that on income should have the opposite sign. However, this test can not prove that those coefficients should be equal in magnitude, since the proxy of the income variable can not represent the individual monthly income. In addition, this article introduces another variable which can be a proxy of outdoor water use. That is lot size showing the effect on water demand. The last variable used in the model, house size,does not have much effect on water demand and is dropped out in the final model.