Study On Prediction Of Demand For Construction Land In Qianjiang City

As industrialization process accelerating urbanization level, the demands of construction land are also increasing rapidly. The constant expansion of the scale of construction land in China will perpetuate the contradiction of more people and less land more prominent. Therefore reasonable construction land demand forecasting is particularly important. It is in this background that we research construction land total demand of the Qianjiang City from 1997 to 2005 and related social economic statistics to find suitable forecasting methods to predict future construction land demand of the Qianjiang City.(1)Research background and significance of research are described in the introduction section, through the analysis of the present situation both at home and abroad and the current situation and existing problems of construction land-use about the Qianjiang City, this paper puts forward the objectives of the study and the major content and the technical route.(2)The second part introduces construction land demand forecasting theory basis. It mainly includes the meaning of construction land, land supply scarce theory, rational growth theory, system engineering theory and the relation between human and land theory.(3)The third part focuses on analyzing construction land demand forecasting model, including the multivariate linear regression model, gray GM (1,1) model and the BP neural network combination forecast model. These three models is not only the core of this paer, but also the base of construction land demand forecasting of the Qianjiang City.(4) the fourth part is regional overview about the Qianjiang City, mainly introduces the natural condition, social economic condition and land use status, etc and summarizes the problems existing in the construction land use.(5) the fifth part is the core part of this paper. First it introduces the multivariate linear regression model and model test and the correction of optimization model, and establishes multiple linear regression equation to predict the demand for the Qianjiang City construction land. Secondly it by DPS6.85 constructs gray GM (1,1) model and the model test, the conclusion is that this model can be applied directly to long-term forecast construction land. Then it introduces the BP neural network model, including the process of the construction of the neural network model and study training, and gets neural network fitting residual graph and fitting values of the total amount of construction land from 1998 to 2005. Last it evaluates the prediction result for the three models, and grey GM (1,1) model fitting effect is superior to other two models and forecasts construction land demand of the Qianjiang City by grey GM (1,1) model.(6) the last chapter includes conclusion and the insufficiency, conclusion summarizes the research results, the insufficiency pointed out the insufficient place of the paper.