本文選題：房地產 + 預警　； 參考：《蘭州交通大學》2013年碩士論文

【摘要】：迄今為止，中國房地產行業(yè)經歷了30多年的發(fā)展歷程，其發(fā)展模式實現了從計劃體制到市場制度的轉變，但無論房地產行業(yè)在發(fā)展中處于何種階段和運行模式，始終離不開國家、政府的調控行為。1998年以前，國家對住宅實行福利分房的計劃經濟制度，房改以后，住宅作為房地產行業(yè)的主體結構走向了市場，其發(fā)展迅猛，市場運行劇烈波動，體現在價格波動巨大，投資、投機活動頻繁等，直接或間接的影響著國民經濟的運行和穩(wěn)定，因此加強房地產市場的監(jiān)管對經濟社會中的各個角色尤為重要。 在上述背景下，本文開展了關于房地產市場預警系統建模與分析的研究，首先閱讀了大量文獻，了解到國內外有關房地產市場監(jiān)管和預警的現狀，著重介紹了幾個典型代表國家的房地產市場監(jiān)管體系和國內房地產預警研究的進展，為找到合理的預警方案奠定了基礎；其次，本文闡述了房地產周期理論、房地產預警理論和要素、房地產市場波動成因理論，然后將經濟學原理和房地產行業(yè)相結合，為下文指標的選取等后續(xù)工作奠定了理論基礎。 本文最終選取的是基于神經網絡的房地產市場預警手段,首先介紹了預警建模需要的準備工作，即數據預處理，包括采用時差分析篩選指標和警情警度的數值定義、區(qū)間劃分；其次，選取BP神經網絡算法，詳細介紹了算法原理和應用算法建模的各個步驟環(huán)節(jié)，實現了BP神經網絡與預警系統建模分析相結合,為實證分析的進行做了原理性論述。 本文選取了天津市房地產市場為樣本進行了實證分析，首先介紹了選取天津市作為樣本城市和劃分指標時間區(qū)間的依據，描述了天津市房地產行業(yè)的發(fā)展歷程；其次，通過查詢了天津市統計年鑒和天津市統計局網站，獲得了天津市房地產預警指標的詳細數據，確保了數據的真實性和準確性；最后，對數據進行數據分析和處理，篩選了天津市房地產市場預警指標，運用BP神經網絡進行建模，借助MATLAB軟件編程，，實現了BP神經網絡的訓練和參數的確定，并預測了2012年天津市房地產市場的警情，得出了市場運行為“熱”的結論。 文章最后，總結了本文的研究成果和結論，著重分析了預警系統中的不足，對未來的研究方向進行了展望，對今后預警研究的發(fā)展從制度角度提出了一些政策性建議。
[Abstract]:Up to now, China's real estate industry has experienced more than 30 years of development, its development model has realized the transformation from the planning system to the market system, but no matter what stage and operation mode the real estate industry is in the development, Before 1998, the state implemented the planned economy system of housing welfare and divided housing. After the housing reform, housing as the main structure of the real estate industry went to the market, and its development was swift and violent. The fierce fluctuations in market operation are reflected in the huge price fluctuations, frequent investment and speculative activities, which directly or indirectly affect the operation and stability of the national economy. Therefore, strengthening the supervision of the real estate market is particularly important to the economic and social roles. Under the above background, this paper has carried out the research on the modeling and analysis of the real estate market early warning system. First of all, it has read a lot of literature and learned about the current situation of the real estate market supervision and early warning at home and abroad. This paper mainly introduces several typical real estate market supervision systems and the progress of domestic real estate early warning research, which lays the foundation for finding a reasonable early warning scheme. Secondly, this paper expounds the real estate cycle theory. The theory and elements of real estate early warning, the theory of cause of real estate market fluctuation, and the combination of economic principle and real estate industry lay a theoretical foundation for the following work, such as the selection of indicators. This paper finally selects the real estate market early warning means based on neural network. Firstly, the paper introduces the preparation work needed for early warning modeling, that is, data preprocessing, including the numerical definition of time difference analysis screening index and alarm degree, interval division; Secondly, the algorithm of BP neural network is selected, the principle of the algorithm and the steps of applying the algorithm modeling are introduced in detail. The combination of BP neural network and early warning system modeling and analysis is realized, and the principle of the empirical analysis is discussed. This article selected Tianjin real estate market as the sample to carry on the empirical analysis, first introduced the Tianjin city as the sample city and the division index time interval basis, described the Tianjin real estate industry development course; secondly, By querying Tianjin Statistical Yearbook and Tianjin Bureau of Statistics website, the detailed data of Tianjin real estate early warning index are obtained to ensure the authenticity and accuracy of the data. Finally, the data are analyzed and processed. The early warning index of Tianjin real estate market is screened, the model is modeled by BP neural network, the training and parameter determination of BP neural network are realized by MATLAB software, and the warning situation of Tianjin real estate market in 2012 is predicted. The conclusion that the market is running as "hot" is concluded. Finally, this paper summarizes the research results and conclusions of this paper, focuses on the analysis of the shortcomings of the early warning system, prospects for the future research direction, and puts forward some policy suggestions for the future development of early warning research from the perspective of the system.
【學位授予單位】：蘭州交通大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：TP183;F299.23

【參考文獻】

相關期刊論文 前2條

1 彭翊;城市房地產預警系統設計[J];中國房地產;2002年06期

2 趙黎明,賈永飛,錢偉榮;房地產預警系統研究[J];天津大學學報(社會科學版);1999年04期

相關碩士學位論文 前2條

1 余健;南京市房地產市場預警系統模型及其應用研究[D];東南大學;2004年

2 裘建國;基于神經網絡的南京市房地產市場預警系統研究[D];東南大學;2006年

本文編號：1923952