A Study of Forecasting for Real Estate Price in Taipei area
碩士 === 明志科技大學 === 工業工程與管理研究所 === 100 === This study is mainly to investigate the trend and structure of average housing price in both Taipei city and New Taipei city, Taiwan. The raw data is purchased from Market Bulletin of Taiwan's real estate transaction, which is a collection of more than 1...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2012
|
Online Access: | http://ndltd.ncl.edu.tw/handle/87039886687576694189 |
id |
ndltd-TW-099MIT00030005 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-099MIT000300052015-10-13T21:01:53Z http://ndltd.ncl.edu.tw/handle/87039886687576694189 A Study of Forecasting for Real Estate Price in Taipei area 大台北房地產價格預測之研究 Hsu,Yin Chieh 許胤捷 碩士 明志科技大學 工業工程與管理研究所 100 This study is mainly to investigate the trend and structure of average housing price in both Taipei city and New Taipei city, Taiwan. The raw data is purchased from Market Bulletin of Taiwan's real estate transaction, which is a collection of more than 10 real estate agencies. Range of obtained data is from the January of 2004 to the June of 2009, averaged monthly. In this study, three major data types of pre-owned houses (residence, commercial office, and store) are analyzed and compared. Total of 16 independent variables are selected through a variety of previous research, including leading indices, simultaneous indices, and lagging indices. Predictions of future housing prices are made by both Back-propagation Neural Networks and Support Vector Regression, with three different variable selection procedures. It is our goal to provide a relatively better prediction of Taipei City and New Taipei City real estate prices, hoping to supply a lower risk decision support for buyers. The research finding shows that: (1) Both support vector regression analysis and neural network performed better in predicting housing price of residence. (2) The predictive accuracy of support vector regression analysis is higher than BPNN. (3) Applying proposed trial and error method as variable selecting procedure is better than the others. (4) Predictions of commercial offices and stores are not satisfactory, while the prediction of residence housing price is quite accurate. The selected major variables are: the real estate price with one month lag, the price of new estate case, the benchmark prime lending rates, construction stocks index, money supply, consumer price index, architectural changes in the amount of loan balance, gross domestic product, households annual growth rate, salary of construction employees ,building transfer the registration number, buildings construction permit area, and new house loan amount influence forecasting model in Taipei metropolitan’s real estate prices model. Hong-Yuh Lin Kuen-Tai Chen 林鴻裕 陳琨太 2012 學位論文 ; thesis 108 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 明志科技大學 === 工業工程與管理研究所 === 100 === This study is mainly to investigate the trend and structure of average housing price in both Taipei city and New Taipei city, Taiwan. The raw data is purchased from Market Bulletin of Taiwan's real estate transaction, which is a collection of more than 10 real estate agencies. Range of obtained data is from the January of 2004 to the June of 2009, averaged monthly. In this study, three major data types of pre-owned houses (residence, commercial office, and store) are analyzed and compared. Total of 16 independent variables are selected through a variety of previous research, including leading indices, simultaneous indices, and lagging indices. Predictions of future housing prices are made by both Back-propagation Neural Networks and Support Vector Regression, with three different variable selection procedures. It is our goal to provide a relatively better prediction of Taipei City and New Taipei City real estate prices, hoping to supply a lower risk decision support for buyers. The research finding shows that: (1) Both support vector regression analysis and neural network performed better in predicting housing price of residence. (2) The predictive accuracy of support vector regression analysis is higher than BPNN. (3) Applying proposed trial and error method as variable selecting procedure is better than the others. (4) Predictions of commercial offices and stores are not satisfactory, while the prediction of residence housing price is quite accurate.
The selected major variables are: the real estate price with one month lag, the price of new estate case, the benchmark prime lending rates, construction stocks index, money supply, consumer price index, architectural changes in the amount of loan balance, gross domestic product, households annual growth rate, salary of construction employees ,building transfer the registration number, buildings construction permit area, and new house loan amount influence forecasting model in Taipei metropolitan’s real estate prices model.
|
author2 |
Hong-Yuh Lin |
author_facet |
Hong-Yuh Lin Hsu,Yin Chieh 許胤捷 |
author |
Hsu,Yin Chieh 許胤捷 |
spellingShingle |
Hsu,Yin Chieh 許胤捷 A Study of Forecasting for Real Estate Price in Taipei area |
author_sort |
Hsu,Yin Chieh |
title |
A Study of Forecasting for Real Estate Price in Taipei area |
title_short |
A Study of Forecasting for Real Estate Price in Taipei area |
title_full |
A Study of Forecasting for Real Estate Price in Taipei area |
title_fullStr |
A Study of Forecasting for Real Estate Price in Taipei area |
title_full_unstemmed |
A Study of Forecasting for Real Estate Price in Taipei area |
title_sort |
study of forecasting for real estate price in taipei area |
publishDate |
2012 |
url |
http://ndltd.ncl.edu.tw/handle/87039886687576694189 |
work_keys_str_mv |
AT hsuyinchieh astudyofforecastingforrealestatepriceintaipeiarea AT xǔyìnjié astudyofforecastingforrealestatepriceintaipeiarea AT hsuyinchieh dàtáiběifángdechǎnjiàgéyùcèzhīyánjiū AT xǔyìnjié dàtáiběifángdechǎnjiàgéyùcèzhīyánjiū AT hsuyinchieh studyofforecastingforrealestatepriceintaipeiarea AT xǔyìnjié studyofforecastingforrealestatepriceintaipeiarea |
_version_ |
1718053050293157888 |