我国房地产定价模型研究

发布时间：2018-03-31 09:28

本文选题：房地产　切入点：商品房销售价格　出处：《武汉纺织大学》2013年硕士论文

【摘要】：自1998年我国对住房实物分配制度停止后，我国商品房销售价格就开始不断上涨，国家对商品房的价格和销售情况不断进行调控，但是房地产的价格仍在快速上涨，它已经极大地影响了我国居民的生活和整个国民经济的平稳发展，商品房价格已经成为了一个被广泛关注的社会问题和经济问题。影响房地产价格的因素错综复杂，目前对商品房销售价格影响因素的分析方法也多种多样，本文拟采用定性分析和定量分析相结合的方法进行研究。影响因素的分析很重要，而房地产的定价更是受到关注。对于房地产的定价，目前有很多种方法，针对影响因素分析和房地产定价的问题，本文主要采用多元线性回归模型、GM(1,1)模型、BP神经网络模型和灰色神经网络模型分别进行研究，并进行模型的比较和选择。本文具体分为以下六个部分。第一部分为绪论，介绍了本文研究的背景与研究意义，提出了问题，并叙述了本文的研究思路与方法，指出了本文的重难点、创新与不足。第二部分为综述，在对部分有代表性的参考文献进行研究、分析和比较的基础上，分别介绍了国内外对相关问题的研究现状。第三部分对房地产价格的影响因素进行了分析，首先对商品房平均销售价格影响因素进行定性分析，主要分为内部影响因素和外部影响因素，内部影响因素则包括了成本因素和营销目标，而外部因素则分别从经济环境因素、社会环境因素、行政因素和其他因素等四大因素进行分析，，然后在定性分析的基础上，选择了本文认为最相关的9种因素进行多元线性回归的定量分析，利用Eviews6.0软件，对多元线性回归模型进行逐步回归，消除因素间的多重共线性，最后确定了竣工房屋造价和全国人口总数为对我国商品房平均销售价格影响最大的因素，并肯定了定量分析与定性分析的结果的一致性。第四部分对商品房定价进行数学建模，分别利用GM(1,1)模型、BP神经网络模型和灰色神经网络模型对我国商品房销售价格进行模拟与预测，通过模型模拟的检验，确定了三种模型在我国商品房定价中的适用性。第五部分为模型的比较与选择，主要通过模型自身的优缺点比较和与实际的结合来对模型进行选择，并作出了多元线性回归模型适合于确定影响因素的分析，GM(1,1)模型适用于样本量少的高精度的短期预测，BP神经网络适用于大样本的中长期预测，而灰色预测模型则适合于样本量少的高精度的预测。第六部分是对本文的总结和展望。
[Abstract]:Since the cessation of the system of real estate distribution in our country in 1998, the selling price of commercial housing in our country has been rising continuously. The price and sales of commercial housing are constantly regulated by the state, but the price of real estate is still rising rapidly.It has greatly affected the life of Chinese residents and the steady development of the whole national economy. The price of commercial housing has become a social and economic problem that has been widely concerned.The factors affecting the real estate price are complicated, and the analysis methods of the factors affecting the sale price of commercial housing are also varied. This paper intends to use the qualitative analysis and quantitative analysis to study the factors.The analysis of influencing factors is very important, and the pricing of real estate is paid more attention.There are many methods for real estate pricing at present. Aiming at the problem of influencing factors analysis and real estate pricing, this paper mainly adopts the multivariate linear regression model (GM1 / 1) to study the BP neural network model and the grey neural network model, respectively.The model is compared and selected.This article is divided into the following six parts.The first part is the introduction, which introduces the background and significance of this study, puts forward the problems, describes the research ideas and methods, and points out the important difficulties, innovations and shortcomings of this paper.The second part is a summary. On the basis of the research, analysis and comparison of some representative references, this paper introduces the current research situation of related issues at home and abroad.The third part has carried on the analysis to the real estate price influence factor, first carries on the qualitative analysis to the commodity house average sale price influence factor, mainly divides into the internal influence factor and the external influence factor.Internal factors include cost factors and marketing objectives, while external factors are analyzed from four major factors, namely, economic environmental factors, social environmental factors, administrative factors and other factors, and then on the basis of qualitative analysis,The 9 factors considered most relevant in this paper are selected for quantitative analysis of multivariate linear regression. By using Eviews6.0 software, the multivariate linear regression model is gradually regressed to eliminate the multiple collinearity between factors.Finally, it is determined that the cost of the completed house and the total population of the country are the most important factors affecting the average selling price of the commercial housing in China, and the consistency between the quantitative analysis and the qualitative analysis is confirmed.The fourth part carries on the mathematics modeling to the commercial housing pricing, respectively uses the GMX1) model and the grey neural network model to carry on the simulation and the forecast to our country commercial housing sale price, passes the model simulation test.The applicability of the three models in the pricing of commercial housing in China is determined.The fifth part is the comparison and selection of the model, mainly through the comparison of the advantages and disadvantages of the model itself and the combination of the actual model to select the model.The multivariate linear regression model is suitable for determining the influencing factors. The model is suitable for the short term prediction with small sample size and the BP neural network is suitable for the medium and long term prediction of large sample.The grey prediction model is suitable for high precision prediction with small sample size.The sixth part is the summary and prospect of this paper.
【学位授予单位】：武汉纺织大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F299.23;F224

【参考文献】