But such changes take place over the years: – five – ten years. This suggests that at least a first approximation, the estimated corrections are constant. A local causes that influence the pricing, do not depend on time. Slow changes in valuation adjustments reflected in the fact that different classes of housing becoming more expensive or cheaper in many ways, but this difference is the second-order correction. For example, the 2003 level of prices in Moscow grew by 36%. And depending on the class housing various apartments went up by 30% – 40%. This is the second-order correction – a weak dependence of the local causes from time to time.
It is more significant when the overall price level remains roughly constant, but is all but erased amid strong growth or decline in prices. The second group of factors that affect the pricing – it's global causes. They associated with macroeconomic parameters such as level of economic development and business in the city, income levels and, in general, living in this city. As well as its status and prestige. It is these reasons determine that the level of housing prices (as on most other goods) in Moscow, higher than St. Petersburg and St. Petersburg than in Novgorod and . And the difference in the prices of similar properties located in different cities, also roughly proportional to each other.
This allows us to compare the general price level in the same city as the price level in another, and to assert that the prices of similar apartment in various cities will be roughly proportional to the ratio of price level in those cities. In this case, the price level is common for this city. It is the same component in the price of each apartment is not dependent on its properties and parameters. Although would be correct to say that the price of each apartment depends on the general price level – reflecting the impact of global economic factors, and a set of adjustments to it – taking into account the effect of local housing properties. The very same level of Price is only a function of time and does not depend on all the local parameters. All of the above, we can express the symbolic formula: Ck (t, pi) = G (t) + Lk (pi) In this formula, Ck (t, pi) – price per unit area (square meters), separate the k-th apartment in given time t. Local parameters of the apartment described by a set of values of pi – is the location of the apartment, house type, in which it resides, kitchen area, the presence or absence of balconies, etc. According to the above ideology, this price is made up of two components. Function G (t) describes the contribution of global macroeconomic factors and just a very general level of prices in the city – one for all objects. The values of Lk (pi) represent the contribution of local differences. They are different for each of the k-th apartment and depend on the set of its characteristics. As a first approximation, we consider corrections Lk (pi) does not depend on time. Mathematically, this approach achieves separation of variables. In this model, the first term describes the dependence of the overall price level in the city from time to time. Equally important to the dynamics of the market analysts. The second term – estimated adjustments for teachings of the price of the accommodation. This is the essence of the method of describing the real estate market, developed and applied analytical center irn.RU