Spatial analyses in price models: looking for evidence from a land price study
It is well-known that empirical models to estimate determinants of price are generally ad hoc in nature, not least because markets for different goods are affected by different contexts and factors. Even if the good is the same, data and measurement constraints often yield different model specificat...
Saved in:
| Main Author: | |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/38624/4/ICRMMS_spatial.pdf http://irep.iium.edu.my/38624/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | It is well-known that empirical models to estimate determinants of price are generally ad hoc in nature, not least because markets for different goods are affected by different contexts and factors. Even if the good is the same, data and measurement constraints often yield different model specifications (and hence, results) in different studies. Model selection is therefore, a particularly important exercise, one that is hoped to reveal a ‘final’ model which best subscribes to market realities and available data. For a good as heterogeneous as land, where every plot exhibits unique combination of attributes, price studies usually employ the hedonic modeling approach. With the advancement of spatial econometrics and computing abilities, more price studies are incorporating varying forms of spatial analyses to test structural stability of the price function or to control for spatial error dependence and serial auto-regressive effects on price. Estimation of a hedonic price function using Malaysian dataset of agricultural land sales values indicates the presence of spatial disaggregation and spatial dependence. However, diagnostic tests and actual estimation of spatial models are not unambiguous while predicted errors do not seem to vary all that much from ones generated by simpler models. Despite the conceptual appeal of spatial analyses, the inefficiency attributable to spatial biases might not be large enough to cause critical errors in policy decisions. |
|---|