Today, the Analyzes Department of has completed the estimation of the econometric model of the value of commercial (non-residential) real estate in the Warsaw region.

A total of 2,404 properties in 10 districts were described in detail from maps, plans and documents, and 2404 properties in 10 counties (and even a few% more, because some fell out during the vision). After rejecting garages, small share transactions, low Visa transactions and real estate with demolished (development) buildings, 1,670 properties were adopted for the model estimation. After excluding outliers / influencers, the final estimate was made on the basis of 1,480 properties from the period from the beginning of 2006 to October 2013.

During the construction of the model, the biggest problem was the strong correlation between some features, which required the use of techniques such as Principal Component Analysis (PCA).

Ultimately, the following market features were taken into account in the model:
– date,
– type of right (ownership / UW),
– share,
– poviat (Warszawa, Grodziski, Warsaw West, Mińsk, Pruszkowski, Otwocki, Piaseczyński, Legionowski, Nowodworski, Wołomin), – distance from Warsaw, from the poviat and from the commune in km,
– distance from the voivodeship road or better in km,
– category of public road (DKA – national road (motorway), DKS – national road (expressway), DK – national road (other), DW – provincial road, DP – district road, DG – municipal road),
– access from a public road (direct entry from the road, indirect entry from the road, entry via an internal road, unregulated entry, entry through a building complex, entry through an easement),
– total area of ​​buildings (PC = PZ x Lcond.),
– basement (% PC),
– land area,
– shape of the property (Wk),
– type of building location (frontage, detached, in a building complex),
– number of buildings –
number of functional parts,
– crossing the property border (none, minor, significant),
– historicity (0-1),
% share of architectural aesthetics (0-5),
% share of technical condition (0-7),
% share of standard ( 0-5),
% share of modernity (0-4),
% share of construction (solid, light, openwork, partially without walls, without walls),
% share of technology (prefabricated, brick, steel, wooden),
% share of construction years (2006+, 2000-05, 1990s, pre-war, 80s, 70s, 60s, 50s),
% share of space types (BA – office and administration, OK / SZ – culture and health , UH – service and commercial, no information, GH – catering and hotel, GP – garages / parking lots, W – workshop, M – warehouse and storage, SB – social and living, MU – residential and service, WZ – shelters / roofs, P – production, G – economic, IN – other, RS – agricultural specialist, R – agricultural general).
The terms of the transaction were also taken into account: type of seller and buyer (public, private), tender, forced sale.

In the future, additional features that have been specified for the properties in the database may be taken into account.

Tomasz Kotrasiński

head of the Research Department at