在過去房地產的相關研究文獻中，有著許多指出房屋價格波動與總體經濟情況有 著關聯性和因果關係，以及利用房屋特徵值對房屋物件進行大量估價的研究，本 篇論文的目的，是期望尋找是否可以選取總體經濟變數，對在配對方法下所預估 出來的房屋時間效果，進行建模預測，在模型中，我們假設影響被解釋變數影響 解釋變數的係數，並非恆常不變，而是會隨著時間而有所不同，於是在模型中加 入變數自我迴歸的假設，做為模擬未來期係數的主要方法。此篇論文所採用的研 究資料，是利用 2012 年上線的實價登錄系統之交易資料，研究對象包含：台北 市、新北市、桃園市、新竹市、高雄市，在經過配對方法計算出平均時間效果率 後，將此時間效果視為房屋價格的平均成長率作為被解釋變數和預測目標。

Many thesis point out that the growth rate of housing price have cause effect relationship with the conditions of economics. Following the effort done by Shi et al (2012) figuring out the cause effect relationship between the housing price and economics variables exist in the main cities in Taiwan, in this thesis we try to put several economics variables in model from Hui et al (2010) and add autocorrelation characteristic to simulating the growth rate of housing price. We use the transition data of actual price registration which established since August 2012 as research materials. Taipei City, New Taipei City, Taoyuan City, Hsinchu City and Kaohsiung City are the cities involved in this thesis. By using matching approach, we withdraw the time effect from transition records. We build a forecasting model with the usage multi-external variables as external variables and the assumptions of autocorrelation. Key word: Actual Price Registration, Matching Approach, Multiple Regression, Forecasting Model, Autocorrelation.

1. Introduction 1
1.1 Motivation 1
1.2 Purpose 2
2. Literature Review 3
2.1 The matching approach 3
2.2 The cause effect between macroeconomics index and housing price 3
2.3 The assumption of linear model and valuation 4
3. Methodology 4
3.1 The matching approach to withdraw the time effect 5
3.2 The linear model assumption of ATE 5
3.3 The autocorrelation model of coefficients 6
3.4 The comparison between simulated ATE and real ATE 6
4. Data 6
4.1 The time effect of housing price 6
4.2 The macroeconomics index variables 13
5. Numerical result 16
5.1 The numerical result of regression model 16
5.2 The numerical result of simulated coefficients 17
5.3 The comparison between the real and simulated ATE 17
6. Conclusion 23

Dickey, D.A. and W.A. Fuller, 1979, Distribution of the estimates for autoregressive time series with a unit root, Journal of the American Statistical Association, 74, 427-431
Liu, 2016, Using a matching approach in New Taipei City’s real estate market, Department of Quantitative Finance National Tsing Hua University Master Dissertation
Sam K. Hu et al. 2010. A Hierarchical Bayesian Approach for Residential Property Valuation: Application to Hong Kong Housing Market. International Real Estate Review.
McMillen, D., McDonald, J., 2002. Land values in a newly zoned city. The Review of Economics and Statistics 84, 62–72
McMillen, D., 2008. Changes in the distribution of housing prices over time: structural characteristics, neighborhood or coefficients? Journal of Urban Economic 64, 573–589.
McMillen, D.P., 2010. Price indices across the distribution of sale prices: a matching approach.
McMillen, D.P., 2012. Repeat sales as a matching estimator. Real Estate Economics, 40(4), 745-773
Rapach, D.E. and J.K. strauss.2009. Differences in housing price forecastability across US states, International Journal of Forecasting, 25, 351-372
Sam K. Hu et al. 2010. A Hierarchical Bayesian Approach for Residential Property Valuation: Application to Hong Kong Housing Market. International Real Estate Review.
史育菱、徐莞婷、張禮韻等人，(2012)，總體經濟變數與房地產價格波動之 Granger 因果關係研究- 以臺北市、新北市、臺中市、臺南市、高雄市中古屋交 易市場為例，逢甲大學土地管理學系第廿七屆畢業論文
林秋瑾、王建安、張金鶚(1997)，房地產景氣與總體經濟景氣於時間上領先、同 時、落後關係之探討；國家科學委員會研究會刊：人文及社會科學,七卷,一期
洪淑娟、雷立芬(2000)，中古屋、預售屋／新成屋房價與總體經濟變數 互動關 係之研究；臺灣銀行季刊第六十一卷第一期


25

陳彥光（2010），房地產景氣與總體經濟變數之關聯性分析－價與量的對比 分 析，東吳大學經濟學系碩士論文。
陳力維 (2001) 台灣房地產價格變動因素之研究，淡江大學財務金融學系；碩士 論文
彭建文、張金鶚（2000），總體經濟對房地產景氣影響之研究，國家科學委員會 研究彙刊；人文及社會科學，第十卷，第三期，頁 330-343
楊奕農 (2009) 時間序列分析：經濟與財務上之應用；第二版，台北：雙葉書廊