多元回归和神经网络在武汉市房地产预测中的应用比较

发布时间：2018-02-27 21:24

本文关键词： 房地产市场预测多元回归神经网络 BP模型　出处：《华中科技大学》2013年硕士论文　论文类型：学位论文

【摘要】：房地产是一个复杂的综合性系统工程，关系的国民经济的命脉。房地产市场自身存在周期性波动规律，过大的波动幅度不利于房地产市场的持续稳定健康发展，进而影响到这个国民经济的可持续发展。房地产市场预测正是对房地产相关指标波动幅度的预测，，建立科学的房地产预测系统，可以有效地防止房地产市场的非正常波动，从而促进房地产市场的健康、稳定发展。这也就是本文的研究目的所在。本文首先是介绍相关的预测模型，多元回归模型和神经网络模型。包括这两种模型的原理、算法和检验，以及神经网络模型中本文主要应用的BP模型。其次，分别利用多元回归模型和BP模型在预测和模式识别领域的成熟运用，以武汉市房地产市场为对象进行实证研究，并运用spss、Matlab等统计数据挖掘工具根据武汉市房地产市场相关数据建立相应模型，重点探讨基于BP神经网络理论的房地产预测的模型和方法，同时对两种模型的房地产预测模型和方法分别进行了分析。最后，通过对两种不同的预测方法做了预测分析比较后，得出比较优质的模型，形成一个综合预测分析系统，并以此为依据提出相关政策调控建议。通过对预测结果的分析，发现它基本符合武汉市房地产发展的实际情况，这表明本文所建立的房地产预测模型是有意义的，本文理论分析充分，所做的预测分析有一定实用价值。
[Abstract]:Real estate is a complex comprehensive system engineering, which is the lifeblood of the national economy. The real estate market has its own periodic fluctuation law, which is not conducive to the sustained, stable and healthy development of the real estate market. This will affect the sustainable development of the national economy. The real estate market forecast is precisely the prediction of the fluctuation range of the related indicators of real estate. The establishment of a scientific real estate forecasting system can effectively prevent the abnormal fluctuations of the real estate market. In order to promote the healthy and stable development of the real estate market, this is the purpose of this paper. This paper first introduces the related prediction model, multivariate regression model and neural network model, including the principle, algorithm and test of these two models, as well as the BP model which is mainly used in the neural network model. Using the multivariate regression model and BP model in the prediction and pattern recognition field of mature application, taking Wuhan real estate market as the object of empirical research, Using SPSS Matlab and other statistical data mining tools to establish the corresponding model according to the relevant data of Wuhan real estate market, focusing on the real estate forecasting model and method based on BP neural network theory. At the same time, the real estate forecasting models and methods of the two models are analyzed separately. Finally, after comparing and forecasting the two different forecasting methods, a better model is obtained and a comprehensive forecasting and analysis system is formed. Through the analysis of the forecast results, it is found that it basically accords with the actual situation of the real estate development in Wuhan, which shows that the real estate forecasting model established in this paper is meaningful. The theoretical analysis in this paper is sufficient, and the prediction and analysis made in this paper has certain practical value.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F299.23;F224;TP18

【参考文献】