This article considers the issue of identification of the key factors determining the composition and classification of housing in a large city. The results of the analysis conducted made the authors conclude that the structure and classification of housing are primarily determined by the level of housing availability for local residents; it depends on the city population count, the average cost of 1 sq.m. of residential housing, and the average income level. The authors developed a multifactor linear regression equation that explains the 98.3% correlation between the parameters mentioned and can be used to forecast housing market development in cities. Based on the performed analysis of the key parameters of residential buildings that are being built in Krasnoyarsk, the authors provided recommendations for apartment parameters to be used when designing apartment blocks.