The rapid changes of environment have a considerable effect on the real estate prices. The consequences of urban constructions (like new tramways or metro stations, new bridges or industrial areas) are durable and difficult to measure. The estimation of these environmental transformations is important as it has a strong effect on the local real estate market. The methodologies aimed to solve this problem are based on the construction of sector house price indices that help to evaluate and in some cases, even to foresee the possible changes of the real estate situation in each sector. The construction of indices focuses on measuring the evolution of effective real estate values in time basing on an assumption that we possess information on the representation samples for all the transactions concerned. There exist different methodologies for the construction of price indices in real estate: some of them are based on the price of transactions, others present expert evaluations or quotas of real estate societies. The objective of our research is to present an original methodology for house price indice construction in real estate based on the repeat sales model. We intend to show that the introduction of time dependency into this model allows creating annual sector indices susceptible to price differences provoked by major environmental changes.