基于灰色人工免疫算法的房地产行业供需模态分析和预测

发布时间：2018-05-28 15:40

本文选题：房地产 + 多元线性回归　；参考：《武汉理工大学》2014年硕士论文

【摘要】：房地产行业是国民经济的支柱产业之一，近年来房地产行业发展迅猛，为国民经济的发展做出了巨大的贡献，在我国的经济社会发展中有着举足轻重的地位。同时，商品房的供需及价格与人民的生活息息相关，房地产问题已经成为最重要的民生问题之一。当前很多学者对房地产行业的供需及价格问题做了大量的研究，但是大多是从政策、资金等层面的因素来考虑的，定性分析多于定量分析。通过建立数学模型从定量的角度来分析影响房地产行业发展的因素间的数量关系，较准确的预测房地产供需及房价的走势从而为进行有效的调控提供决策支持，是一个值得探索的方向。本文从定性分析的角度深入研究了影响房地产需求、供给及销售价格的因素后，选取中华人民共和国国家统计局《2013年中国统计年鉴》中的相关数据，使用多元线性回归方法对其进行了定量分析。然后使用人工免疫算法及灰色预测理论，针对房地产的供需及房价提出了有效的预测模型。本文的主要研究成果有：（1）在定性分析的基础上结合数据的可获得性，分别选取对房地产需求、供给及售价影响最密切的因素利用逐步回归方法进行了定量分析。针对不同的样本数据，使用了向前逐步回归、向后逐步回归、带常量的逐步回归及不带常量的逐步回归等方法进行建模，并对比了各方法所求得模型的拟合精度和有效性。（2）灰色系统理论着重研究小样本、贫信息的问题，房地产行业的年度供需值及销售价格数列的数据量少，适合使用灰色预测模型进行建模。本文使用GM（1,1）模型对商品房的需求量、供给量及平均销售价格进行了预测。为提高标准GM（1,1）模型的拟合精度，将免疫克隆选择算法引入灰色GM（1,1）模型，提出了两种优化算法。（3）针对标准GM（1,1）模型使用相邻累加数取均值生成的背景值不能准确反映数据序列变化情况，导致预测值与实际值有较大的误差的不足，本文提出了一种使用免疫克隆选择算法对背景值参数进行寻优的改进算法。（4）针对标准GM（1,1）模型使用最小二乘法求解待估参数a、b，而最小二乘法有许多使用限制可能导致求解的预测值精度不高的不足，本文使用免疫克隆选择算法在最小二乘法求得的a、b初始解的基础上继续进行寻优计算，，获得更为精确的a、b值。使用基于免疫克隆选择算法改进的GM（1,1）模型对商品房需求、供给及售价进行建模，实验结果表明两种改进算法都比标准GM（1,1）算法对原始数据序列的拟合精度高，本文提出的灰色人工免疫算法可以对商品房供需及售价的变化趋势进行更准确预测。
[Abstract]:The real estate industry is one of the pillar industries of the national economy. In recent years, the real estate industry has developed rapidly, which has made a great contribution to the development of the national economy, and has played an important role in the economic and social development of our country. At the same time, the supply and demand of commercial housing and its price are closely related to the people's life. The real estate problem has become one of the most important livelihood issues. At present, many scholars have done a lot of research on the supply and demand and price of real estate industry, but most of them are considered from the aspects of policy, capital and other factors, qualitative analysis is more than quantitative analysis. By establishing a mathematical model to analyze the quantitative relationship among the factors influencing the development of the real estate industry from a quantitative point of view, the paper predicts the trend of real estate supply and demand and house prices accurately, thus providing decision support for effective regulation and control. Is a direction worth exploring. Based on the qualitative analysis of the factors affecting real estate demand, supply and sales price, this paper selects the relevant data from the Statistical Yearbook of China 2013 of the National Bureau of Statistics of the people's Republic of China. The multivariate linear regression method was used to quantitatively analyze it. Then, using artificial immune algorithm and grey prediction theory, an effective forecasting model for real estate supply and demand and house price is proposed. The main research results of this paper are as follows: 1) on the basis of qualitative analysis, combined with the availability of data, the factors most closely affecting real estate demand, supply and selling price are selected for quantitative analysis by stepwise regression method. For different sample data, the methods of stepwise regression, backward stepwise regression, stepwise regression with constant and stepwise regression without constant are used to model the model, and the fitting accuracy and validity of the models obtained by these methods are compared. 2) the grey system theory focuses on the problem of small sample and poor information, and the annual supply and demand value of real estate industry and the quantity of data in sales price series are less, so it is suitable to use grey prediction model to model the model. In this paper, the demand, supply and average selling price of commercial housing are forecasted by GM1) model. In order to improve the fitting accuracy of the standard GM-1) model, the immune clone selection algorithm was introduced into the grey GM1 / 1) model, and two optimization algorithms were proposed. (3) the background value generated by using the mean value of adjacent accumulative number in the model can not accurately reflect the variation of data sequence, which leads to the deficiency of large error between the predicted value and the actual value. In this paper, an improved algorithm for optimizing background parameters using immune clone selection algorithm is proposed. The least square method is used to solve the parameters to be estimated, and the least square method has many limitations, which may lead to the low accuracy of the prediction value. In this paper, the immune Clone selection algorithm is used to continue the optimization calculation on the basis of the initial solution obtained by the least square method, and a more accurate value of AGB is obtained. Based on the improved GM1) model based on immune clone selection algorithm, the demand, supply and selling price of commercial housing are modeled. The experimental results show that the two improved algorithms are more accurate than the standard GM1 / 1) algorithm in fitting the original data sequence. The grey artificial immune algorithm proposed in this paper can predict the change trend of supply and demand and price of commercial housing more accurately.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F299.23

【参考文献】