1 |
住戶屬性與實質居住環境水準分析之研究陳建忠 Unknown Date (has links)
鑒於國內住宅需求殷切,同時卻仍有大量空屋無人居住使用,顯示居住環境需求與供給間確存有極大的差異,以致公私部門不能利用有效財源作有效的投資。本研究為縮短供需差距,摒除已往研究偏重所得與需求關係的研究,建購所得型式及居住支出之居住環境需求理論模型,複於需求模型導入住戶規模因素,使所得、支出與住戶屬性並列為需求影響因素。其次,利用住宅現況調查資料,分析居住環境水準與相關住戶屬性之對照關係,並著手分析各居住群組之居住環境需求量,以提升分析可信度,對住戶群落予以細分及區隔居住環境群落,凸顯各種影響因素特性。本研究實證結果:居住支出增加、有業人口數愈多,家計負責人教育程度提高及年齡愈大,則其住宅面積及房間數數量愈多,而且其住宅座落與各種都市服務設施距離相對縮短,但國小、工作場所與市場則無顯著關係。
有關理論模型建購,係基於Stone-greay 函數符合需求模型相關假設,及效用之可分性、可加件原則,以儲蓄率自所得中另離析支出之需求模式,再由個人居住需求累計為住戶居住環境需求模型,進一步設定其最小居住環境需求量為Barten之人口規模函數。在實證分析方面,為了瞭解住戶居住環境水準,以次數分析、關聯分析就歷年發展、地區別、住宅權屬、家計負責人屬性分析其分組水準及分析頻度,並檢定住戶與居住環境的變數關聯程度,以擇定需求量迴歸分析的應變數組及自變數組,及就具有居住支出項的大量樣本進行住戶屬性及居住環境特性的群落分析,俾進行各群組需求分析。
本研究雖已跳脫以所得推導居住環境需求窠臼,惟由於資料及分析係援用政府既成問卷,造成研究領域受限,需再就研究之主題深入設計調查問卷。本研究需求函數係設定為直線,然而居住環境需求量與住戶屬性間若非線性關係時,則其相關係數偏低,且無法驗證兩者間之需求關係。住戶自變數(行業、職業、所得等屬性)間,並未檢定其是否已存在高度相關,無法達到自變數完全獨立之要求。居住環境設施具有共用之基本生活設備時,雖可測定其居住環境水準,但無法進行其需求分析。群落分析固然能分離居住環境群及住戶群,但易使迴歸分析模型內部分虛擬變數與其他變數形成共線而無解,而且本分析僅偏重實質居住環境需求,對於住戶非實質需求、偏好及社會文化群族傾向等因素,在經濟學之需求模型中均無法予以論證,有待識者續以作為研析之題材。 / The domestic housing are in great demand, but on the other hand lots of housing remain vacant. This phenomenon reveaIs there exists a significant difference between supply and demand for housing environments. Therefore, both public and private sectors are unable to make the most use of available funds to invest effectiveIy. The purpose of this study is to lessen the above difference. Prior studies stressed the importance and the relationship between income and the demand for housing. In this study,first the anthor build up a theoretic demand model for housing environments.
This demand model is mainly relevant to no only income but also household expenditure. Secondly, the author converts the factor of household scale into the model. That makes income, expenditur,and household attributes serve as three major factors affecting this demand model. Then, applying data (housing status quo ) gathering from government statistics, the author analyzes the relationships between quantities of housing environments and relevant household attributes. FinaIly,the author analyzes the quanities of living environments for each Iiving cluster,which is specified and segregated, to explicate the property of each factor,thus to enhance the reliability of this study. The result of this study indicates that those household with more living expenditure,more emplyed employed persons, higher education and more age, will have more floor area and room number. In addition, the distances between their residenes and the variety of public service facilities are relatively shoty,but they are litte related with the elementary school, work place and market.
Theoretically,this model has been built based on Ston-geary utility function which is suitable for certain hypotheses for demand model. And frOm the additivity and the separability of this utilty function, the author derives the demand mode, reIevant to household expenditure, for housing environments. Then the author integrates individual demand model into the household one for living environments, and further defines the least quantities for living environments as Barten's population scaIe function. In empirical performance, the author applies frequency analysis and Chi-square analysis to analyze physical Iiving environments,respecting the past 20 yeare, different districts, household tenure, and household attributes. Ih addition. the author examines the co-relations of those variables between dwelling units and household environments to determine dependent and independent variables for regression. Besides, a great deal of samples with household expenditure has been inspected by cluster anaiyis.
Although this paper analyzes the demand function for housing environments on many factors instead of only on income (elasticity, the study is somehow limited since the data acquired from government tatistics.It would be more appropriate if we design a better questionnaire proper to this subject. Also, in this paper,the demand function has been defined as linearity,but if the demand quantities of living environment and household properties develop as a non-linear reationship, then the multiple coefficient of determination appears low,hence the demand relationship can not be tested between them.Moreover, the independent variables for resident themse1ves, such as industry,occupation, and income, have not been tested whether if they are highly related,thus these bariables do not fit in the requirement of complete classified into different living environment Ievels, their demands can not be anaIyzed.
Likewise, cluster analysis can segregate living environment clusters and resident clusters, it is apt tO make some variables, especiaIIy those in those regression models with dummy variables, convert into the combination of other variables thus can not be explained. Ih summary,this study underlines the demand for physical living environments. To those factors, such as non-realistic demand, preference, and social/cultural inclinations, they can not be tested in the demand models of economic theories, nowadays. This challenge stiII needs more endeavors to make.
|
2 |
模糊統計分類及其在茶葉品質評定的應用 / Analysis fuzzy statistical cluster and its application in tea quality林雅慧, Lin, Ya-Hui Unknown Date (has links)
模糊理論開始於 1960 年代中期,關於這方面的研究與發展均已獲得相當不錯的成果.其中尤以在群落分析應用上的專題研究更是廣泛.Bezdek 提出的模糊分類演算法,乃根據 Dunn 的C平均法所作的一改良方法.但仍有其缺點,例如,未考慮權重且以靜態資料為主. 有鑑於此,本研究對 Bezdek 之方法加以改進推廣,提出加權模糊分類法.對於評價因素為多變量時,應加入模糊權重的考量.此外更結合時間因素,使準則函數成為動態的模式,將傳統的模糊分類法由靜態資料轉為動態資料形式,以反映真實
的情況. / Research on the theory of fuzzy sets has been growing steadily since itsinception during the mid-1960s. The literature especially dealing with fuzzycluster analysis is quite extensive. But the research on FCM still has somedisadvantages. For instance, the
|
Page generated in 0.0139 seconds