On the other hand, the housing market gives high importance to some spatial attributes.
Although the results obtained for structural variables can be compared to other studies, the
location attributes analyzed in this work cannot be compared due to the special attributes of the
spatial area investigated. Despite this, the market prefers houses close to the sea, close to
downtown or in districts with better social, economic, cultural, or environmental expectations for
the citizens (Districts 2 and 10). These results could be of interest to city-planning experts and
decision makers. Consumers more value areas close to the seaside and which are capable of
fulfilling their preferences for good public and private services, green areas, recreational areas
and cultural and leisure infrastructures. On the other hand, the results also show that Malaga
citizens prefer to live near the center of town, and so policies to restore and rebuild poorly
maintained areas could be beneficial from many points of views.
Finally, according to our results, public administrations should encourage estimation
studies of house pricing -- and housing attributes -- in different areas using hedonic
methodology. Such research could help to make comparisons over time and between areas on
which to base policy-making for the sector.
NOTES
1 Traditionally, blocks of flats or condominiums in Spain employ a person -the portero- who is
the caretaker and cleans the communal areas of the property.
2 The standardized regression coefficients have been calculated with the expression
where, βist. represents the standardized coefficient of variable i, βi the non-standardized
coefficient, Si the standard deviation of variable i, and Sdep.var the standard deviation of the
dependent variable.
βist = βi ∙
( S'
i/
S
de dep. var. y
3 The statistical confirmation of this result is done by a significance testing of coefficients, where
the null hypothesis means unitary elasticity of the housing price in relation to the surface area.
The statistic t in such a test should take an absolute value ±1.96, at the 5% significance level.
Thus, the null hypothesis is rejected for the presence of unitary elasticity because the value of
the statistic t is = -6.27 at the 95% confidence level.