The database used in this study include a total of 21.517 homes sold during the period2004-2007 by a real estate agency for for which the details concerning the characteristics of the transactions and the location of these are known. Empirical models used in this analysis are, in one hand, hedonic regression, and secondly, decomposition quantílica. The first model determines when the change in the distribution of property prices can be explained by the size, quality and location of the house sold, whether as if there are systematic differences between different cities within the same country . The second model helps answer questions like the evolving distribution of house prices over time or similar. This model examines the effect of covariates on various quantiles of the logistic function of price. Using this model the authors can decompose quantile decomposition changes in the distribution of property prices in the distribution originated by the explanatory variable, on one side, and on the changes induced by the quantilic regression coefficients.n of property prices over time in several Spanish regions. Traditional literature provides very few results and studies on this topic.The aim of this paper is to present a quantilic estimation of housing price index between 2004 and 2007 in several Spanish cities. To perform this analysis, the authors proceed to decompose the change in property prices in parts due to changes in explanatory variables and parts due to coefficients change over time. During the period 2004-2009, the Spanish property price growth of approximately 31.91%. This period can be divided into two smaller sub-sequences: while during 2004-2007 price indices increased approximately 44.31%, in 2007 these prices entered a downward dynamic in which fell about 8.60%. Note that this greater appreciation in the price indices did not occur equally in all Spanish cities. Based on this evidence, the aim of this paper is to understand more satisfactorily the evolution of the distributio
The distribution of prices for homes for sale in Spain in 2007 adopted a less sesgad distribution, with a tendency to move to the right and a smaller kurtosis distribution compared to 2004. All the Spanish cities experienced a change in the distribution of their housing prices, although this change was greater in smaller cities like Valencia or Bilbao compared to Madrid or Barcelona. The difference is greatest in low and high percentiles. We can also determine how much of the difference in the distribution of property prices between 2004 and 2007 is explained by it’s coefficients. The final results suggest that a single price index is not an appropriate measure to measure changes in prices, as the variance of the price index increases over time. Allow assessment ratios vary across quantiles can help us to obtain a more realistic picture of the change in the overall distribution of property prices than an estimator that focuses only on the mean or median. This result comes reinforced by the fact that the variation of the ratio quantile assessment and its components is different between the cities that make up the index.