Examining the vintage effect in hedonic pricing using spatially varying coefficients models: a case study of single-family houses in the Canton of Zurich

Swiss Journal of Economics and Statistics

Table 3 Mean and covariance estimates \(\hat{\omega }_{MLE}\) of the SVC model (5)

Covariates	Mean \(\hat{\mu }_{j}\)	Sign. level	Range \(\hat{\rho }_{j}\)	Variance \(\hat{\sigma }_{j}^{2}\) and \(\hat{\tau }^{2}\)	Sign. level
Intercept	9.1555 (0.1571)	***	13.5511 (4.1157)	0.0413 (0.0116)	***
\(Z.age\)	− 0.1495 (0.0254)	***	18.9370 (41.6414)	0.0009 (n.a.)	n.a
\(Z.age^{2}\)	0.0114 (0.0062)	.	1.7153 (0.8673)	0.0003 (0.0003)	n.s
\(\log \left( {volume} \right)\)	0.5132 (0.0222)	***
\(\log \left( {plot size} \right)\)	0.1720 (0.0129)	***
\(renov\)	0.0274 (0.0059)	***
\(standard\)	0.0911 (0.0082)	***
\(micro\)	0.0385 (0.0075)	***
\(year quarter 20184\)	0.0040 (0.0160)	n.s
\(year quarter 20191\)	0.0235 (0.0137)	.
\(year quarter 20192\)	0.0327 (0.0135)	*
\(year quarter 20193\)	0.0357 (0.0135)	**
\(year quarter 20194\)	0.0279 (0.0149)	.
\(SFH type\) 2	− 0.0062 (0.0120)	n.s
\(SFH type 3\)	− 0.0274 (0.0165)	.
\(energy 2\)	0.0119 (0.0281)	n.s
Error term				0.0242 (0.0010)

The corresponding estimates’ standard errors are given in parenthesis. In most cases, the standard errors can be approximated and computed by the Hessian from the numeric optimization. For the mean and the variance estimates, we use a two-sided Z- and a Wald-test to test whether \(\hat{\mu }_{j} \ne 0\) and \(\hat{\sigma }_{j}^{2} > 0\), respectively. This is only possible if the standard error is available.
Significance levels: ‘.’ p < 0.1; ‘*’ p < 0.05; ‘**’ p < 0.01; ‘***’ p < 0.001; ‘n.s.’ not statistically significant, ‘n.a.’ not available