Different valuation methods and determinants of housing prices in Budapest, Hungary are examined in this paper in order to describe price drivers by using an asking price dataset. The hedonic regression analysis and the valuation method of the artificial neural network are utilised and compared using both technical and spatial variables. In our analyses, we conclude that according to our sample from the Budapest real estate market, the Multi-Layer Preceptron (MLP) neural network is a better alternative for market price prediction than hedonic regression in all observed cases. To our knowledge, the estimation of housing price drivers based on a large-scale sample has never been explored before in Budapest or any other city in Hungary in detail; moreover, it is one of the first papers in this topic in the CEE region. The results of this paper lead to promising directions for the development of Hungarian real estate price statistics.

For cultural and aesthetic reasons, it seems obvious, that sustaining the high quality of historic buildings is a necessity. There are a number of organizations maintaining historic buildings by a monitoring system all over Europe, although there is no published data on the economic advantages of the practice. The aim of our research is to show that besides the cultural and aesthetic arguments, there are sound economic reasons for continuous maintenance. Our study focuses on the costs by comparing the case of regular maintenance, to that of isolated renovation that takes place every 15 or 20 years after a long period of negligence. In our pilot we have monitored six typical historic buildings to identify the economic facts alongside the aesthetic and cultural arguments, in order to clarify the importance of keeping our built heritage in good condition. Keywords historic buildings • continuous maintenance • regular monitoring • cost effectiveness Acknowledgement This work is connected to the scientific programme of the "Development of quality-oriented and harmonized R+D+I strategy and functional model at BME" project. This project is supported by the New Hungary Development Plan

There are plenty of historic buildings bearing different stylistics in Budapest and many of them have residential function. In the city center of Pest, most of the properties are historic buildings constructed between the period of the Austro-Hungarian Compromise in 1876 and the World War II, but Buda also has some residential dwellings with historic value. Estimation of the value of the Budapest residential housing is an important issue for owners, real estate developers and investors, nevertheless not many studies have focused on the value components of those buildings in Central Eastern Europe or Hungary. In this paper the value components of Budapest residential flats were identified using the hedonic regression method. On a sample of more than 1800 residential properties of Budapest the differences between historic, panel and other buildings were compared. The conclusion can be drawn that altering aspects are relevant for each segment. Even the categories determine large differences between panel buildings and non-panel buildings regarding the value. For the historic properties, the existence of balcony, the up-to-date type of heating, the good condition of the flat, the unique panorama, the location in Pest City, the vicinity of parks and the distance from noisy facilities are the most important factors. Meanwhile for panels the allocation on lower floors, the better heating system, the good condition, the location in Buda and the vicinity of market are the factors that have the major positive effect on the value. For the non-historic and non-panel buildings the balcony, the up-to-date heating system, the good condition, the luxurious Buda district location, vicinity of parks and remoteness of noisy facilities are the most important components of value.