Skip to main content

Table 4 Robustness error-correction model results for residential real estate development

From: On real estate development activity: the relationship between commercial and residential real estate markets

  Model 1 Model 2
\(\hbox {Ln}(\hbox {C}_{\mathrm{jt}}^c )\) 0.39*** 0.18*
(0.13) (0.10)
\(\hbox {Ln}(\textit{GDP}_t )\) 0.39* 0.49*
(0.26) (0.26)
Constant 1.47 2.37
(1.57) (1.57)
N 60 60
F test for weak instruments 12.75 5.20
Test for exogeneity \(<\)0.01 0.01
(Critical value) (3.84) (3.84)
  1. We used the 2sls estimation approach to test models 1 and 2. We further tested the robustness of this 2sls procedure and also tested with GMM. For the F test for weak instruments, we rely on the minimum eigenvalue statistics. For model 1, we find that the instrument is not weak, whereas, for model 2, we find that the instrument is weak at the 5 % level. To test for exogeneity, we used the Durbin Wu-Hausman test. The score test does not reject the null hypothesis of exogeneity and, therefore, we consider commercial real estate development as exogenous at the 5 %. Robust standard errors are given between brackets. Significance at the 1 %, 5 % and 10 % levels are denoted by ***, ** and * respectively