【摘要】：土地供應計劃是科學系統(tǒng)利用土地資源的基礎,建立長期有效的土地供給體系,才能實現(xiàn)房地產(chǎn)市場的長期穩(wěn)定和繁榮。而土地供應量的預測是土地供應計劃的技術支撐,只有科學合理地預測土地供應量,才能使制定的土地供應計劃切實有效地調(diào)節(jié)房地產(chǎn)市場,維持房地產(chǎn)市場長期穩(wěn)定發(fā)展。 本文借鑒國內(nèi)外相關研究方法,在全面分析天津市房地產(chǎn)市場的運行狀況的基礎上,系統(tǒng)地梳理了“十一五”期間天津市土地供應、商品房交易以及主要的社會經(jīng)濟指標的統(tǒng)計數(shù)據(jù),應用統(tǒng)計分析模型、BP神經(jīng)網(wǎng)絡模型對天津市“十二五”期間逐年土地供應量進行了預測。 本文首先界定了經(jīng)營性房地產(chǎn)用地需求與供給預測的基本思路和在預測時應遵循的原則。然后,根據(jù)易獲取可量化的原則分別建立影響住宅和商業(yè)房地產(chǎn)需求量、供給量的因子體系,并采用灰色關聯(lián)度模型篩選出對因變量影響較大的關鍵因子：基于回歸模型理論,利用matlab軟件的曲線擬合工具對遴選出的關鍵因子做非線性回歸擬合,得到單因子在“十二五期間”逐年的預測值,為住宅和商業(yè)房地產(chǎn)BP神經(jīng)網(wǎng)絡主模型預測打下基礎。 本文采用具有三層結構的BP神經(jīng)網(wǎng)絡模型對住宅和商業(yè)房地產(chǎn)需求量、供給量進行預測：將“十一五”期間住宅和商業(yè)房地產(chǎn)的數(shù)據(jù)輸入到網(wǎng)絡分別進行訓練學習,根據(jù)網(wǎng)絡性能誤差確定合適的中間層神經(jīng)元數(shù)目；然后,將回歸擬合得到的“十二五”期間住宅和商業(yè)房地產(chǎn)數(shù)據(jù)輸入到訓練好的網(wǎng)絡,得到住宅和商業(yè)房地產(chǎn)需求量、供給量預測值。 住宅和商業(yè)房地產(chǎn)用地預測值可由住宅和商業(yè)房地產(chǎn)需求量、供給量與平均容積率通過運算得到。平均容積率可由天津市“十一五”期間中心城區(qū)、濱海新區(qū)和郊區(qū)縣容積率的平均值及各區(qū)域“十二五”期間經(jīng)營性房地產(chǎn)用地的比例,綜合得到；再由住宅和商業(yè)房地產(chǎn)需求量、供給量預測值通過容積率就可得到住宅和商業(yè)房地產(chǎn)用地量的預測值。 為確保得到的住宅和商業(yè)房地產(chǎn)用地預測值更加合理,更加具有指導性意義,本文根據(jù)天津市“十一五”期間住宅和商業(yè)房地產(chǎn)的累計竣工銷售狀況,按照系統(tǒng)科學總量平衡的原則進行去存量修正,以使未來住宅和商業(yè)房地產(chǎn)市場土地供需平衡。最后,給出對天津市“十二五”期間住宅和商業(yè)房地產(chǎn)土地供應計劃的建議。
[Abstract]:Land supply plan is the basis of scientific system to utilize land resources and establish a long-term effective land supply system to realize the long-term stability and prosperity of the real estate market. The forecast of land supply is the technical support of land supply plan. Only scientific and reasonable prediction of land supply can make the land supply plan regulate the real estate market effectively and maintain the long-term stable development of the real estate market. Based on the comprehensive analysis of the operation of the real estate market in Tianjin, this paper systematically combs the land supply in Tianjin during the 11th Five-Year Plan period by referring to the relevant research methods at home and abroad. Based on the statistical data of commercial housing transaction and main social and economic indicators, the statistical analysis model and BP neural network model are used to predict the land supply year by year in Tianjin during the 12th Five-Year Plan period. This paper first defines the basic ideas and principles to be followed in forecasting the demand and supply of commercial real estate land. Then, according to the principle of easy to obtain and quantifiable, the factor system of influencing the demand and supply of residential and commercial real estate is established, and the key factors which have great influence on dependent variables are screened out by using the grey relational degree model: based on regression model theory. By using the curve fitting tool of matlab software, the selected key factors are fitted by nonlinear regression, and the predicted value of single factor is obtained year by year during the 12th Five-Year Plan period, which lays the foundation for BP neural network main model prediction of residential and commercial real estate. In this paper, the BP neural network model with three-layer structure is used to predict the demand and supply of residential and commercial real estate. The data of residential and commercial real estate during the 11th Five-Year Plan period are input to the network for training and learning. Determine the appropriate number of interlayer neurons according to the network performance error; then, input the regression fitting data of residential and commercial real estate into the trained network, and get the demand for residential and commercial real estate. Supply quantity forecast value. The predicted value of residential and commercial real estate land can be obtained by calculation of demand, supply and average volume ratio of residential and commercial real estate. The average volume rate can be obtained from the average volume rate of the central urban area, Binhai New area and suburban counties during the 11th Five-Year Plan period of Tianjin and the proportion of commercial real estate land in each region during the 12th Five-Year Plan period. According to the demand of residential and commercial real estate, the forecast value of supply quantity can be obtained through the volume ratio of residential and commercial real estate. In order to ensure that the forecasted values of residential and commercial real estate land are more reasonable and more instructive, this paper, according to the cumulative sales situation of residential and commercial real estate in Tianjin during the 11th Five-Year Plan period, In order to balance land supply and demand in future residential and commercial real estate market, destocking correction is carried out according to the principle of systematic scientific aggregate balance. Finally, the suggestions on the land supply plan of residential and commercial real estate in Tianjin during the 12th five-year plan are given.
【學位授予單位】：天津師范大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：F299.23

【參考文獻】

相關期刊論文 前10條

1 劉永平;房地產(chǎn)需求模型研究[J];重慶工學院學報;2001年01期

2 張洪;;我國城市土地供應政策的經(jīng)濟分析——以云南為例[J];財貿(mào)經(jīng)濟;2007年06期

3 張慶佳;林依標;;土地有效供應對房價的影響分析[J];福建論壇(人文社會科學版);2008年04期

4 臧俊梅;王萬茂;;我國土地利用的經(jīng)濟效益初探[J];廣東土地科學;2005年04期

5 鄒高祿;成都市商業(yè)房地產(chǎn)市場需求敏感性因素分析[J];資源與人居環(huán)境;2004年05期

6 丁成日;城市土地需求分析[J];國外城市規(guī)劃;2005年04期

7 婁文龍,朱青;土地供給方式對房地產(chǎn)發(fā)展之影響[J];湖北廣播電視大學學報;2004年04期

8 蔡育天;上海土地供應總量及其調(diào)控對策[J];上海綜合經(jīng)濟;1998年07期

9 高麗坤;李淑杰;王錫魁;蔣甫玉;;土地供應市場與房地產(chǎn)市場關系的研究[J];華中農(nóng)業(yè)大學學報(社會科學版);2006年01期

10 呂連生 ,程龍干;中國城鎮(zhèn)新增用地供給量的研究[J];經(jīng)濟研究參考;1996年F1期

，

本文編號：2118723