52 Motivations behind Property Developers’ Price Cut: “Profit-Sharing” vs. “Sales” 表3 Trace Test and Maximum Eigenvalue 共整合檢定結果 Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob.** None * At most 1 * 0.357354 0.102401 31.9117 6.265827 20.26184 9.164546 0.0008 0.1712 Hypothesized No. of CE(s) Eigenvalue Max-Eigen Statistic 0.05 Critical Value Prob.** None * At most 1 * 0.357354 0.102401 24.6435 6.265827 15.89210 9.164546 0.0011 0.1712 註：Trace Test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level ** MacKinnon-Haug-Michelis (1999) p-values 表4 VECM 模型落後期數決定結果 Lag LogL LR FPE AIC SC HQ 1 2 3 4 5 6 -513.8265 -507.6599 -506.0606 -504.1928 -503.1716 -502.1654 6.096147 11.32234* 2.831477 3.184519 1.674046 1.583559 86499.81 80614.91* 87329.65 93869.45 103877.0 115198.7 17.04349 16.97246* 17.05117 17.12108 17.21874 17.31690 17.25112 17.31850 17.53563 17.74396 17.98004 18.21662 17.12486 17.10807 17.24103 17.36519 17.51710 17.66951 註：* Indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion （三）議價率與 30 天銷售率之時間關係及調整速度 為進一步釐清議價率與30 天銷售率的短期調整及長期均衡關係，本文以公式 (1) 建立VECM，主要係實證議價率及30 天銷售率為具有長期關係之序列，及在短 期波動相互調整的過程。表5 中，本文由議價率及30 天銷售率之共整合關係、誤 差修正項及係數向量 (α)，推導VECM；在表6 之Granger 檢定中，證明議價率及 30 天銷售率為落後2 期並互為因果關係，即議價率會Granger 影響銷售率，而銷售 率亦會Granger 影響議價率。
RkJQdWJsaXNoZXIy ODg3MDU=